2nd Crash in 6 months

Discussion in 'Cinestar 8' started by David N Atkisson, Apr 18, 2013.

  1. David N Atkisson

    Joined:
    Nov 24, 2012
    Messages:
    73
    Likes Received:
    0
    Hey guys,

    I never thought I would have to post another one of these but I had another disastrous crash with my Cinestar 8 today. I was flying over the beautiful Palm Springs Desert shooting our Convention Center and on my return to home my copter went down. If any of you remember my posts about the first one then this one is worse. At least from my initial inspection of the remains. Here is a picture of the actual wreckage.

    IMG_0653sml.jpg


    And here is me holding my poor bleeding baby!!!

    IMG_0655sml.jpg

    I was flying at pretty far out, and got a warning on the mx-20 that my telemetry was dropping out so I started coming back home immediately. I flow with Alt Hld on which I normally don't do but because of my distance I decided to be safe and click it on. then on the way back she started to descend and would not respond to my controls. This was in my opinion only a stretch for my skills because it was around 2 football lengths out from me and I don't normally fly that far away. Otherwise it was a straight shot out and back with nothing fancy for me to do except to fly out and back with no obstructions or any other things to hinder me.

    Here is a link to the video of the raw footage if anyone cares to comment.



    I also have flight data from the navi board and the graupner if anyone would like to help me decipher the data.
     
  2. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    Oh no! So sorry.
    If you can post the GPX file from that flight, maybe we can help you understand what happened, and maybe why.
     
  3. David N Atkisson

    Joined:
    Nov 24, 2012
    Messages:
    73
    Likes Received:
    0
    I attached the gpx file of the flight for all who are interested. Also the password for the video if it asks you is "password" . we decided to not put crash footage out for general consumption considering the nature of our Industry at this time. No need supplying any ammunition for negative reactions. Thanks guys.
     

    Attached Files:

  4. Gary Haynes

    Gary Haynes Administrator
    Staff Member

    Joined:
    Jul 6, 2012
    Messages:
    5,211
    Likes Received:
    460
    David I'll look at the GPX file tonight. Any chance that you have the mkp file of your copter settings? Would like to see your come home setup. Also post the Graupner file. There is a program that read those files also.

    And I know the feeling. Was just stunned when my Hoverfly Pro locked up and crashed. Video was interesting. Had a GoPro, NEX7 running on board and was recording the FPV feed. Really helped show what went wrong.
     
  5. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,369
    Likes Received:
    1,162
    Hi David:
    Oh crap! (British Understatement).

    Here's the summary from MK_GPXTOOL:
    Comments to follow when I've looked through the file.

    Andy.


    MK Version: FC HW:2.1 SW:0.88m + NC HW:2.0 SW:0.28o​

    Flight date: 2013-04-18 12:35:46 PM​
    Flight time: 12:35:46 PM - 12:38:13 PM (147 secs, 00:02:27)​
    Batt. time : 149 secs, 00:02:29​

    Elevation(GPS) : 1.028 29.17 36.828 m (min/avg/max)​
    Altitude(Barom.): -1.1 30.77 38.45 m​
    Vertical speed : -3.65 0.13 4.78 m/s​
    Max speed : 89.8 km/h​
    Max target dist.: 0 m​

    Sats : 7 8 10​

    Voltage : min. 13.4, max. 16.5 V​
    Current : 33.5 69 172 A​
    Wattage : 516 1057 2304.8 W​
    Capacity: 2843 mAh​

    Motor1: 1.4 8.0 25.4 A Temp: 37 57 77 °C​
    Motor2: 2.3 8.0 25.4 A Temp: 35 54 72 °C​
    Motor3: 1.5 7.1 19.2 A Temp: 28 44 53 °C​
    Motor4: 2.0 8.3 16.5 A Temp: 32 53 61 °C​
    Motor5: 2.6 8.8 19.1 A Temp: 37 67 84 °C​
    Motor6: 3.0 9.2 16.6 A Temp: 39 73 93 °C​
    Motor7: 4.0 8.8 24.0 A Temp: 32 60 75 °C​
    Motor8: 3.9 8.1 21.5 A Temp: 32 61 75 °C​

    Magnet Field: 78 79 84 % <- check!​
    Magnet Inclination: 36 56 62 deg​

    Errors / warnings:​
    Compass is disturbed by magnetic fields!
    Error "ERR: GPS lost" (21) occured 9 times!
    Error "Motor restart" (23) occured 2 times!
    FC-Flag "LowBat" (5) occured 7 times!


     
  6. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,369
    Likes Received:
    1,162
    The GPS Lost errors occur in second 2-10 of the flight.
    The Low Bat occurs in the final seconds of the flight: Seconds 138-145.
    The Motor Restart (which indicates BL-Ctrl problem) occurs in the final two seconds of flight: Seconds 144-145.

    You were in Altitude Hold from second 31 of the flight all the way until second 144 (the last recorded data is second 145 -- when Alt Hold does not appear to be on).

    According to the GPX file, things go badly wrong in second 128 when the C8 is a 36 meters. Check the Motor Currents for seconds 128 onwards -- they are shown in units of 0.1 Amps so 36 is 3.6Amps.....you'll see that the motors suddenly start pulling 13.9, 20.5 amps.
    The total current draw is 46 Amps in second 127, but by second 129 is at 110.1 Amps!!

    As you can see in second 128 the LiPo voltage is 15 volts. but this current draw drags it down to 13.6 volts.

    What is not immediately obvious to me, based on just a cursory review is what actually failed.

    When the C8 started to go down, did you start to put in some fairly extreme stick motions in Nick? I'll try and take some more time to look at this but maybe Gary and others will see something that my cursory examination has missed.

    Andy.

    Dave Atkisson Crash Data Second 114-145.png
     
  7. Gary Haynes

    Gary Haynes Administrator
    Staff Member

    Joined:
    Jul 6, 2012
    Messages:
    5,211
    Likes Received:
    460
    David

    Preliminary look.

    At 129 seconds your power draw spikes to 110.1A. Motor 1 and 2 jump from 3.0 to 13.9 and 20.5A respectively. This continues for 5 seconds. This is the same point that the copter starts back towards the takeoff point on Hermosa Dr. Distance was about 1,349 feet. Altitude was 34.4m. The signal RCQuality looks good.

    Things look like they went back to normal for the 5 seconds then got really bad at 138 seconds. Total draw jumps to 134A and peaks at 172A. All of the motors are drawing high amps.

    Looks like loss of control at 144 seconds. Copter is pitched down 128 degrees and rolled to the right nearly 90 degrees. GPS is lost, Low Battery alarm is triggered, motor restart is triggered.

    The copter never went into failsafe mode nor was come home triggered.

    You do have a compass warning that it is is being disturbed by magnetic fields. The Magnet field value was 84% and below which is not generally an acceptable value. Sometimes this value is low from the location of the Navi board to other onboard electronics, wiring or other transmitters. When you rebuild this should be checked. I think the warning is shown when the value is below 90%, 100% or greater is the target.

    The other item I don't think I have ever seen before is that the Available Motor Power in the last two seconds dropped from 255 to 40. Maybe Andy can figure that one out.

    The other item for Andy is the No Serial Link line in the graphview status flags screen. I've seen that on a lot of files yet I don't have it on any of my own GPX files. Not sure what that indicates.

    From the video, towards the end, the motors sound like they are going to nearly full power yet the craft is going down.

    So here's a SWAG for what it's worth. Based on the time of day, you are over a reflective building, blacktop street and the proverbial dusty field, all of which sail plane pilots like due the thermals that come off of them I would hazard a guess that you got caught not by the updrafts but in downdrafts from a thermal.

    Did you see any dust devils around, swirling debris, light stuff in the air? Outside temp? All good indications of up/down air mass movement.
     
  8. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,369
    Likes Received:
    1,162
    I think the impact occurred at that point in the last two seconds of the log.

    The No Serial Link means that you're not connect to MK Tool either via wire (unlikely in this case!!) or Xbee/Bluetooth.

    Not a bad hypothesis but I didn't see any corresponding turbulence in the flight path -- the C8 was coming in fast and low, but completely level. So I'm not sure it was a victim of "The Giant Hand" (that pushes a sailplane down out of the sky).

    Also, I didn't see any signs of anything blowing in the winds that would normally be associated with said Giant Hand.

    I'm leaning more to a failure of the Flight Controller board -- but that's hunch, nothing more. I think the motor restart was perhaps a symptom of the crash, not the cause.

    That said, I'm really speculating...there is nothing specific in the evidence that I've seen that indicates FC failure.

    Andy.
     
  9. Gary Haynes

    Gary Haynes Administrator
    Staff Member

    Joined:
    Jul 6, 2012
    Messages:
    5,211
    Likes Received:
    460
    Loss of connection between the FC and the PDB? Molex connector became disconnected?
     
  10. Jean Gabriel Taboada

    Joined:
    Jan 21, 2013
    Messages:
    72
    Likes Received:
    2
    do you have hot glue on the molex and nav, gps cables? so sorry to see this, been there, its not a pretty feeling.
     
  11. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,369
    Likes Received:
    1,162
    I would have expected to see some I2C errors if that had happened.
    Of course the GPX parameters are only recorded once a second, so a transient happening in between those times would go unrecorded.

    It's nagging at me that the flight path was straight and level....but with the motor currents going nuts as the FC tried to maintain altitude....

    Andy.
     
  12. Jean Gabriel Taboada

    Joined:
    Jan 21, 2013
    Messages:
    72
    Likes Received:
    2
    I had a connector come loose, the copter flipped on its head and crashed but the gpx file showd as if it where a hard landing, no trace of the flipping on its head whatsoever. no i2c error either. i attached the gpx so you see what a normal, slow flight and then a sudden flip on its head and crash looks like on a gpx file.
     

    Attached Files:

  13. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    David,

    I just replayed the flight from the GPX file step-by-step. It's like using a debugger to step through program code.

    You said "on the way back she started to descend and would not respond to my controls", but I don't see any significant loss of altitude until a few seconds before the crash. At that time, the copter went from drawing 58.6 amps to 134 amps, and eventually 172 amps just a second or two before impact. The rapid descent seems to have happened coincidentally to the high power draw, and then the copter crashed. But it began drawing high power at between 34 and 38 meters in altitude.

    I have a couple of thoughts. Others will undoubtedly add theirs.
    1. Depending on your battery (it's age, health and number of duty cycles), I wonder if the high power draw caused the battery to fail.
    2. The copter could have suffered a structural failure of some sort as a result of wear and tear, or (more likely) the violent maneuvers. It went from mostly level to 15° to almost 30° of bank very quickly, and that could have caused something to come loose. A sudden shift in the Center Of Gravity (like it the camera slid on the boom) could account for the copter's rapid attempt to recover equilibrium. I've had just a slightly rough landing cause the booms in the gimbal to slide pretty alarmingly.
    Obviously none of this is going to replace your copter. I'm so sorry, and I (and all of us) appreciate you sharing your dreadful experience.
     
  14. Brad Meier

    Brad Meier Active Member
    Staff Member

    Joined:
    Aug 14, 2012
    Messages:
    749
    Likes Received:
    66
    It seems to me, just by watching the video, your horizontal speed exceeded the lifting ability of the copter. In altitude hold, when you start to return, your speed is say... 10 mph. By the time you impact, the horizontal speed is much greater. At some speed your copter is not able to provide enough power to hold altitude. I have found that speed to be between 30-35 mph just by filming with cars at those speeds. Of course this also depends on motors, weight etc. I suspect you had increased the pitch angle attempting to return faster which in turn caused the FC (in altitude hold) to give more power to maintain level flight. At some point level flight, even at full power, could not be sustained. In this case it seems altitude hold may have be relied upon but it's capabilities were exceeded. Speed at impact should be able to support my theory
     
    Cedar Beauregard likes this.
  15. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,369
    Likes Received:
    1,162
    I just came to the same conclusion, David.

    Let me explain the basis:

    When your C8 is getting too far away from you, you were probably feeling uneasy (I know I do), and so you pull back on Nick to bring it home. I see this happening at second 135 (the RC Stick for Nick starts becoming increasingly negative -- almost half the total excursion possible). The C8 obliges (the Nick angle also goes negative -- nose up).

    The ground speed also increases rapidly -- it starts coming towards you at a ground speed of up to 89 km/hour - with Nick nose-up of 30-40 degrees.

    As it is doing this the center of lift is so far off vertical, that the copter is not generating enough lift to maintain altitude, so it starts to descend. But AH is on, so the FC board does the only thing it can and applies an increasing amount of throttle -- hence the sudden increase in power/current to the motors.

    However, there is no way the copter can maintain altitude at that nose up attitude, so it descends into the ground at a ground speed of about 73 km/hour.

    I suspect to you it must have felt like the C8 was coming towards you fast but losing altitude. At second 140, you suddenly applied almost full throttle. But it's too late, I'm afraid.

    This is what the evidence suggests anyway. Does it seem feasible to you?

    Sorry if we sound like the bearers of bad news....
    Andy.
     
  16. Shaun Stanton

    Shaun Stanton Active Member

    Joined:
    Oct 27, 2012
    Messages:
    748
    Likes Received:
    127
    Ouch that is painful to watch:(

    Because you also had altitude hold on the entire time means that going to full throttle would not have made difference other than commanding it a climb, at a fixed rate that your aircraft was unable to perform at the pitch angle with out backing off on the pitch stick, especially carrying a 5D on board. If you would have done that you would have started a rapid climb. The reason that the battery dropped so low was because of the rapid response to attempt a max climb rate at 40mph the copter attempted to give it its all. The the discharge rate went past safe limits. The electrical load that you were pulling in the last seconds caused the voltage to substantially drop, to try to supply the current required by the speed controllers. A speed controller started to go off line due to load shedding due to a brownout condition. You reached a point where the voltage was not pushing the current fast enough. I agree with Andy's assessment it looks like the aircraft was just flown beyond limits.

    One the things that I find that helps is having a pilot nose camera coupled to the MK-Smart OSD. This gives you status of the craft, lets you see a dot on where your home location is. You can see if you are bringing it back because there is LOS to home position distance number, so you can see if the copter is coming back. You get all of your Sat information so you can see if you have enough sats to bring it back on RTL. I am not sure why you had GPS errors at the start.

    This aircraft can provide a parallax illusion and make you think its not coming back when it is. I have had a few panics where I thought that the AC was at a pitch angle of 45 degrees by staring up at it in the sky without a reverence. I almost panicked but I looked at my OSD and found that it was level. Sometimes looking at it in the sky can give you one of those mind trick illusions if your perspective forces your eyes focuses on the wrong thing.

    Might I also suggest getting a smaller bird to practice on. You can get a DJI F550 kit for arround $500. Replacement parts are cheap. I find it a good platform to practice on. It allows for you to get more comfortable doing different things with multirotors without making 10,000 dollar mistakes. I have crashed my F550 by over temping the ESC's, at the end of the day the replacement parts were about $60 bucks and cleaning off sand.

    Sorry for the loss on the bird, never a good day when that happens.

    Shaun
     
  17. Duane Bradley

    Duane Bradley Active Member

    Joined:
    Dec 6, 2012
    Messages:
    212
    Likes Received:
    26
    In essence then, a power on stall. Lift could no longer manage the burden of weight and began to descend. The faster down it moves the more power needed to counter and reverse the descent - which not enough power was available.

    I'd say the physics of this makes sense and the GPX backs it up.
     
  18. Jason Smoker

    Jason Smoker Active Member

    Joined:
    Oct 23, 2012
    Messages:
    849
    Likes Received:
    67
    another question was it set in vario mode or standard altitude hold? If it was set in the standard mode if it heli starts to drop it creates the ceiling if you want to go higher you have to reset the switch. Correct me if i wrong here?
     
  19. Shaun Stanton

    Shaun Stanton Active Member

    Joined:
    Oct 27, 2012
    Messages:
    748
    Likes Received:
    127
    After much research yeah that is what it looks like. Its called height limitation control. They do not do a good job explaining it. It sets an altitude that the copter can not go above. There is a way to change it with a second poti. I don't think he had it in that mode. It looks like he had it on vario mode.
     
  20. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    From all the data I have kept track of, most crashes involves compass errors and not calibrating on location. I would bet my last dollar based on the compass data and errors shown in the GPX file, that the copter was not calibrated for its location. Now when I say, not calibrated I mean to say you could have attempted it but it could possibly be that it didn't go through. Let me explain.

    If you don't hear the DA DA beep at the end it doesn't calibrate. This happened to me at a local golf course last week when I was 20 miles of where I last calibrated it. I thought the calibration process took but it didn't. I think I rotated it in the wrong direction. I have caught myself rotating the copter east instead of north after I rotate boom 1 east in the process. I didn't hear the da da at the end but didn't heed my own warnings of it. When I got the bird in the air I heard the lady told me GPS loss. I also had stability issues in AH and PH. I landed the bird, recalibrated the compass (this time correctly), and I then heard the DA DA. After that I double checked the GPX files on location and everything was perfect. No errors, no warning from the lady, and stability was smooth as butter. I had 60 minutes of excellent flying.

    I'm not saying that this caused the crash because it seems that the cause of the crash has been already figured out by the brilliant guys on this board. But the compass issues could have added to the cause of the crash because the bird may not have responded fast enough or well enough to safely correct for the dire situation you ended up in.

    I can't state this enough that you have to make sure the bird's compass calibration goes through correctly at the start of flying of each location and its a very good practice to get the bird up 15 feet in the air for 2-3 seconds and get it right back down to check the GPX data. If there's a problem you will see it in the compass values right away. If you don't see anything you can have a little bit of peace of mind that everythings good. It helps calm me down at least.

    A couple other things that I would like David to confirm.
    Did you move your pilot RX to the top of the battery try with the Quadrocopter bracket? If you haven't it really helps improve reception strength the further you get the bird away from you. Also did you make sure that your Pilot TX was not receiving telemetry? Did you bind it to 2 receivers like has been documented? Just making sure you took care of these details as well for future flying.
     

Share This Page