Weird Iris+ behavior.. Hard landing

So I finally got all my OSD issues resolved and went flying! Within 30 seconds I was on the ground though and I think my Tarot T-2D is toast...

I have a log, I've tried to analyze it but I can't get the auto-analyze tool in MP to read my file. Not sure if it matters or not but I'm on a Surface Pro 3 running W8.1. Log file is attached. 

I've seen some of you look at these an repeat almost to the minute detail what happened to a person flying... I immediately wrote notes about what I thought happened after the event  and I'll share those below. Can someone help me figure out how to read a log file and maybe offer some insight into what may have  happened? 

High level timeline of events...

Powered up TX (AltHld)

Plugged in battery

Couldn't get GPS lock with 10-12 sats showing on the TX, did not have MP running to analyze HDOP, waited approx 6-8 minutes (yellow LED flashing)

Unplugged battery

Powered down TX

Switched to Stabilize mode on the TX

Powered up TX

Plugged battery back in

Got blue LED on Iris+ within a few minutes

Switched back to AltHld on TX

Armed Pixhawk

Armed motors

Applied throttle relatively quickly due to slight breeze and location of launch

Got Green light (GPS lock)

Applied more throttle

Experienced extreme vertical acceleration

Throttled down quickly

Extreme rate of descent

Tried to throttle up, no apparent response

Hit ground hard

Tried to disarm motors (left stick down and left) no response, motors continued to spin at high RPM for approx 10-12 seconds

Centered left stick left-to-right and still all the way down

Moved left stick to left most position again still all the way down

Motors disarmed after approx 2-3 seconds

Disarmed Pixhawk

Unplugged battery

Turned off TX

Gimbal appears to be not responding or going through leveling sequence although I get the normal amber light and blinking blue light.Thankfully the my GP Hero 4 appears to be okay. I'll upload video in a few.

Very perturbed... I was planning to go film some footage at a state park this weekend. 

Anyone have any ideas?

Recent updates...

Installed minimOSD recently. Flew twice in the backyard with no signs of any issues

5/27/2015

Updated Mission Planner to 1.3.27 build 1.1.5620.38044

Performed Radio Calibration after re-centering trim on the radio

Updated to AC V3.2.1

Performed Compass Calibration

5/28/2015

Flew approx 1.5 batteries at nearby park. Great performance from Iris+ until the second landing. Would not disarm motors, had to put foot on the top of the quad until to keep from flipping while holding the left stick down and to the left. 

Also noticed during the first flight I did not have normal OSD information on my monitor. It continuously read "CHARACTER UPDATE". 

5/29/2015

The above sequence of events. 

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

Join diydrones

Email me when people reply –

Replies

    • Thanks Michael. I'll double check but pretty sure I loaded 3.2.1.
    • Pre-Arm checks were turned back on after I did the updates mentioned on 5/27. And I couldn't arm the copter even with 12 Sats so something was preventing it from arming. 

      I recorded all the min/max values for the sticks, two position switches, and pots as2 well as the min/mid/max for three position switches. Throttle range was 991 - 2017 when I did the radio calibration so I'm worried you say it showed 700 as the midpoint. Is that supposed to be 1700?

      When you say firmware is out of date I'm assuming you mean of Copter installed on the Pixhawk? I updated to what Mission Planner suggested. Should I have gone out and found a newer version? If so what do you suggest?

      I definitely tried to disarm and can guarantee I had throttle all the way down, I mean buried.

      Your theory, and Jim's clarification below, make total sense. I'll begin using Stab to land or maybe just start using the Land function although I'm guessing it may suffer some of the same issues as AltHld and Loiter since it may use some of the same logic?

    • Comparison of 3.2 and 3.2.1 GyrX on IMU1 and IMU2. Definitley something wrong after the firmware update. 

      AC_3_2_1_Flight.JPG

      AC_3_2_Flight.JPG

      https://storage.ning.com/topology/rest/1.0/file/get/3702701756?profile=original
    • I reckon 3DR has a warehouse full of busted Iris quadcopters as a results if that gyro issue with firmware 3.2.1. I still don't have my replacement back...this can't be good.

    • Good point, sometimes emotion removes logic from the thought process. I was definitely spooled up yesterday and not thinking straight. You're right, if it was an software issue it'd be more wide-spread I'm sure.

      However, I work very closely with software development teams on a professional basis and every time there is an update "something" fails so I am very leery of the fact that I had perfect flights the previous couple of days before the firmware update and began to experience this issue after the update. I even went so far as to write a log about all the things I did when I updated the firmware for fear that something would happen and I would have to "prove" to someone that I did all the correct steps. 

      Based on all the information I read and my understanding today I did everything as I should. I'm just hoping 3DR will help me out. At the very least, if I did do something wrong let me know what it was so I can prevent it in the future. Hopefully if it is a hardware/software conflict they'll replace my my broken parts. 

    • Well today went almost as bad as it could have. Very glad I removed my GoPro 4 from the now dead gimbal. 

      I went out and hooked up Mission Planner, put the TX in Stab and flew for a few minutes, Then launched again from Stab mod, switched to AltHld, then to Loiter and landed fine. Third launch was in Stab mode too, switched to AltHld and flew about 125 feet away from me at an altitude of about 15 feet.Triggered Land Mode and it landed beautifully! Launched in Stab mode and switched to AltHld again and started flying it back to me. All hell broke loose. Because of the previous landing being done in Land mode i walked out to the quad as it descended so I was about a hundred feet from my laptop. After reviewing the log file I see it triggered a "Bad Gyro Health" warning. 

      It's busted up pretty good, nothing that can't be repaired but it'll take a while I'm sure. 

      I read somewhere there was a batch of Pixhawks that had this problem. I don't see any resolutions posted other than 3DR replacing them. 

      Auto-Analysis here:

      Log File C:/Program Files (x86)/Mission Planner/logs/QUADROTOR/1/2015-05-31 19-29-45.log
      Size (kb) 5235.3642578125
      No of lines 71337
      Duration 0:10:51
      Vehicletype ArduCopter
      Firmware Version V3.2.1
      Firmware Hash 36b405fb
      Hardware Type
      Free Mem 0
      Skipped Lines 0

      Test: Autotune = UNKNOWN - No ATUN log data
      Test: Balance/Twist = GOOD -
      Test: Brownout = GOOD -
      Test: Compass = GOOD - mag_field interference within limits (24.97%)
      Max mag field length (562.35) > recommended (550.00)

      Test: Dupe Log Data = GOOD -
      Test: Empty = GOOD -
      Test: Event/Failsafe = FAIL - ERR found: CRASH
      Test: GPS = WARN - Min satellites: 5, Max HDop: 3.35
      Test: IMU Mismatch = FAIL - Check vibration or accelerometer calibration. (Mismatch: 1.83, WARN: 0.75, FAIL: 1.50)
      Test: Parameters = GOOD -
      Test: PM = FAIL - 30 slow loop lines found, max 12.78% on line 955
      Test: Pitch/Roll = FAIL - Roll (-126.91, line 70313) > maximum lean angle (30.00)
      Test: Thrust = GOOD -
      Test: VCC = GOOD -

      Not sure if the fails are a result of slamming up against my house then into the neighbors yard or if they are symptom/cause of the crash.

      Anyone know what the PM = Fail means? 

      I'd use the APM support site but to be honest it seems like if you don't do and say everything the exact perfect way you get reprimanded. I am really not in the mood to listen to someone tell me I didn't follow proper help request etiquette at this point. 

      Log file attached. If anyone sees anything obvious let me know. 

      2015-5-31_Iris_Crash.txt

      https://storage.ning.com/topology/rest/1.0/file/get/3702701442?profile=original
    • Correct logs...

      2015-05-31 19-29-45.bin

    • So I just went back and looked at the logs before i upgraded the firmware. The IMU1 and IMU2 data plots are nearly identical. After the upgrade, not so much. Is there a problem with 3.2.1 that causes this behavior? Data in 3.2. looks fine. 

    • I think Michael is on to something here about switching out of ALT-HOLD when landing. My understanding of the firmware prior to the most recent update is this, and please feel free to correct me if I am mistaken...

      In LOITER mode, you can reposition the quad horizontally and vertically in the normal fashion (pitch, yaw, roll, and throttle). The GPS will then detect the new position and altitude and attempt to maintain that position.

      In ALT HOLD mode, however, while you can still alter the horizontal position, the quad will recall the altitude it was at when you switched into ALT HOLD, and make every effort to maintain that altitude by increasing the throttle as necessary. In some cases, it takes a moment for your throttle inputs to be recognized, so this might allow the quad to touchdown, but the Pixhawk will then realize that it is not where it expected, and it will try to regain that position. This would explain the motors continuing to run.

      My suggestion, at a minimum, would be to switch into LOITER mode, or even all the way to STABILIZE prior to attempting to land. 

      Otherwise, I can empathize with your loss, as I suffered a catastrophic loss last month, and have only returned to the air in the last few days.

  • Any chance you can upload this again?  This one errors out (like you mentioned).

    I saw that one of the errors that you received was that when you tried to switch to Loiter the IRIS couldn't do it for some reason.

    -Mike

This reply was deleted.