My number came up yesterday (crash!)

Discussion in 'Cinestar 8' started by Steve Maller, Oct 20, 2014.

  1. Rod Hartzog

    Rod Hartzog Member

    Joined:
    Apr 16, 2013
    Messages:
    62
    Likes Received:
    9
    So sorry to see that machine dented! I know it'll be better than ever when you've finished the re-build and I'm with you on not being able to see in the fog; I see radar sensors from the automobile industry in our future...:cool:
     
    Steve Maller likes this.
  2. Sebastian Meredith

    Sebastian Meredith Active Member

    Joined:
    Sep 30, 2012
    Messages:
    257
    Likes Received:
    41
    Jeepers! Thank goodness for trees. Sometimes we hate them sometimes we love them. Could have been a real mess if it hit something with concrete holding it together. Sorry to see that Steve. On a lighter side us folk are always looking for a reason to strip our rigs or fiddle with them.
     
    Steve Maller likes this.
  3. Jose Luis Ocejo

    Jose Luis Ocejo Active Member

    Joined:
    Sep 20, 2012
    Messages:
    581
    Likes Received:
    44
    True, but my point is if the navi lights are further apart wold it had made a difference for you in that particular situation or not necessary
    just a thought
    Steve are you sending the motors back to KDE for inspection and cleaning or are you pulling the bells apart to see if there is any debris
     
  4. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    Sorry about the incident but at least nobody was hurt.

    I looked over your GPX file and video and to me a GPS issue in combination with the external compass could be the problem. Let me explain before you instantly dismiss my theory. I know that you were in PH up till the last few seconds and you probably took over manual control when things started going out of control but from someone who has seen a lot of GPS blurps thats would it sounds like to me when I hear that on your video and that's how I've seen it respond. Other interesting notes is that you have the no MK3Mag communication and bad compass error toward the end starting at data point 622. I know that could be caused by the impact if the last few recordings are after impact but I have seen that same issue in my copter several times with the external compass. I have experienced weird quirks on both copters with the same setup that seem to be some type of sensitivity in the area that I have flown. When this has occured and experienced the dreaded GPS blurp I thought it was a loose connection at first. I would land double check the connections and put it back up in the air and it would still occur. I then just unplugged the external compass and ran it with the internal compass and the problem was instantly solved. When I would move locations plug the external compass back in the problem would go away. It's happened to me about 4 different times this summer but I was always able to put it in manual mode and land. I do not believe it was a connection issue as I hot glue both sides fot he cable.

    You mention that you killed the motors but I don't see anything in the logs with the poti sticks graph indicating both sticks in the lower left. At the end of the log the gas goes all the way down but the right stick doesn't move close to a position to kill the motors which leads me to believe that just possibly there's something screwy with the datalogging. If thats the case it wouldn't be the first datalogging issue with the newer electornics. If that actually is the case all bets are off as far as the accuracy of when you first experienced the No MK3Mag communication IMO. According to the log it first polls the error at data point 622 where the copter is said to be 6 meters high still. Could it be in the trees rolling around at that point I guess.

    You mention that you thought the acceleration of the motors started acting up because of severe banking from your response of orientation but at least for me, in flying mode 3, I have never been able to replicate anything like that. I have made severe roll and nick commands many times when testing the stability of my gimbal and my copter has never responded like that. But I have seen severe roll banks when the GPS has gone goofy. Again in all instances I was able to save it by going to manual mode which it appears that you did. So maybe the theory has holes, maybe not?

    Just a few things to think about.
     
  5. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,357
    Likes Received:
    1,161
    The sequence of interest is data rows #624 onwards:
    624 nMK3Mag Communication error code appears : motor currents look more or less normal
    625 Emergency Landing appears in FCFlags2 : motor #2 drops to 0.5 Amps
    626 Bad compass value error code appears : all motor currents are low except motor 1

    By 628 all motors are off.

    Looking at the copter's motion, it looks like 624 is the impact with the tree so that these errors are impact-results, not causal.

    Dave: Remember that the log file is not continuous -- it's a sample every 500 msec. If an event occurs in between samples, you will not see it in the log file. That said, the throttle stick moves from a value of 220 in row 618 to a value of 3 in row 619. (Why is the throttle stick value being shown on P3 on MK_GPXTOOL?). So you're right, Dave, I don't see the Motor Off gesture.

    Steve: How come the copter was heading off really fast towards Hwy 101? Was that a result of the disorientation?

    Andy.
     
  6. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    There are a few other things that stick out to me. Starting at Line 612 to 624 it goes from 6 mph to 46 mph and at line 617 there s a roll angle of 62 degrees but I see nothing in the log to show the roll control going extreme left or right. I don't think that User level 3 can produce a roll angle of 62 degrees even if you tried to go extreme. Am I correct? Then at 623 the RC signal goes south and then it goes into emergency landing and come home at 624. Not sure why it would go into emergency landing without going into come home first for a preset amount of time. All this leads me to believe one of 2 things, a bug in the firmware or a flight control board glitch due to the fog or a combination of both.
     
  7. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,357
    Likes Received:
    1,161
    Good question, Dave. I don't know what the limits are for Parameter Set 3.
    It could also be that at 623 the collision with the tree is in progress and that's why the RC signal goes south.

    Andy.
     
  8. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    Even if the RC signal went south it shouldn't go immediately into emergency landing mode.
     
  9. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    Unlike a lot of crashes we've all seen, my copter stayed intact through the crash, so I have an intact GPX file. One of the things that was lost in the crash, though, was my Graupner GR24 receiver (it was on one of the booms that broke). So when the copter impacted the tree and the radio was torn off, the Flight Control board went into emergency landing mode. That's somewhat reassuring to know. But at that point it was plummeting into the underbrush, so it didn't matter.

    WRT the other points you made, Dave, I am open to the possibility that I experienced a significant GPS problem which led to this. While I admit that I *might have* lost orientation, a GPS glitch could also explain the crash. I was flying relatively close to the elevated approach to the Golden Gate Bridge (well, maybe 100-150m away), and the crash came at the end of a 5+ minute flight, so I can't imagine the bridge would up and dramatically increase its magnetic pull at that point. Just FYI, here's a still from the beginning of the flight, clearly showing what I was describing earlier in this thread about the relative height of the roadway compared to where i was flying, and where the copter went down.

    Screen-Shot-2014-10-22-at-6.33.23-AM.jpg
     
  10. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    Just thought about this last night, as I will be down in that same location shooting (no flying). Any cell towers in the area Steve? I know there are a few radio antenna down the coast where the bridge hits the peninsula, but they are 800mHz range.
     
  11. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    Steve

    After Andy verified that the data points of the error came after impact I'm now not so sure about the GPS glitch theory. I am having a hard time understanding the high bank angle of 62 degrees using flight level 3. I am also not sure why it would go instantly in to emergency landing mode instead of trying to go to come home position for the preset amount of time you determine in MK tools. I can't remember the default time but it should have at least 15-20 seconds of come home before it engages emergency landing. The rate of speed in combination with the unusual high banking sounds like either a moisture issue with the flight control board in the fog, or a failure in the flight control board that just randomly happened in the fog or possibly a firmware glitch.
     
  12. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    This should not matter because emergency landing should not instantly engage if the TX signal doesn't reach the flight control board. It should go into come home first and then go into emergency landing after a preset amount of time. I set mine to 2 minutes and I think the stock value is at least 30 seconds before emergency landing is engaged. The idea is for the MK to come home in close proximity to the TX to regain reception if the reception was blocked by stray RF or physically blocked from obstacles.
     
  13. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,357
    Likes Received:
    1,161
    Steve: Thanks for the clarification re: the RX -- that would certainly explain some of the errors.

    Steve/Dave: Magnetism from a bridge will not affect the GPS signal. The GFS signal is up in the 1.2 and 1.5 Ghz radio frequency range. A bridge's magnetic influence would affect that -- it might "shield" the signal from several satellites low to the horizon, though.
    I've seen GPS units that are attached with nedymium magnets to metal surfaces and they work fine. If they have a compass in them, though, that's a different story. Look at the GPS unit in your iPhone -- and see if it change the apparent position if you bring a neodymium magnet close.

    What I'm still unclear of is why the copter suddenly headed off southwest towards to bridge approach at such high speed?
    Andy
     
  14. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    The metal from the bridge can cause the compass readings to go bad, I've personally seen this when I was filming a series of bridges this summer. The copter briefly started yawing as I got really close to the bridge and then returned to normal as flew over it.

    What affects can mositure on the flight control board do to flight performance? I'm starting to question this.
     
  15. Jose Luis Ocejo

    Jose Luis Ocejo Active Member

    Joined:
    Sep 20, 2012
    Messages:
    581
    Likes Received:
    44
    Going to the disorientation part: I am flying a U5 X8 500 booms I recently installed 2 strips of red and green LED lights at the end of each boom in a 45* down angle the booms also have red and green bright monocote decals on them
    today is overcast gray day here I was flying about 20 ft forward and about 100ft up at LOS with the gray sky I could not see the color of booms nor the LED lights I was using sunglasses because of the brightness of the sky against the black copter so it got me thinking perhaps thats how Steve saw his copter and somehow lost orientation.
    LED and color booms are almost useless some what far away in daylight
     
  16. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,357
    Likes Received:
    1,161

    Now that is certainly true -- and my understanding of the MK code is that the compass (via the Compass Effect) is used to stabilize the yaw axis.

    As to the moisture, it typically only affects anything with high impedance. Rain/fog etc. is distilled water so not terribly conductive. I don't think there are any MOSFETs on the flight controller, though -- and those are the devices most likely to be affected.

    As a precaution, I applied conformal acrylic coating to the FC (not the connectors of course), as well as the NC and the GPS card.

    Andy
     
  17. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    Very helpful discussion, lads. I am looking forward to having a bit of time to re-power everything and make sure all the parts are working. Motors are all going back to KDE for their center shaft upgrades (and testing), so it'll be a week or two before I can do anything.
     
  18. MIke Magee

    MIke Magee Active Member

    Joined:
    Jun 29, 2012
    Messages:
    422
    Likes Received:
    103
    Steve, sorry to hear about this. No one was injured, and hopefully we all learn or are reminded about the importance of visual limitations.

    Thanks for sharing. I always learn stuff on this forum.
    -m
     
    Steve Maller likes this.
  19. Shaun Stanton

    Shaun Stanton Active Member

    Joined:
    Oct 27, 2012
    Messages:
    748
    Likes Received:
    127
    Steve sorry to hear this man. Makes me think that condensation from the fog possibly started to mess with you electronics with the erros you getting. Seemed like a lot of things were going offline. I would echo having large underside lights as a backup. Also I do find having bright florescent red aircraft skin tape helps a bit depending how high you are. I have had instances where if I was flying with the sun in front of me the copter becomes a black silhouette. I am going to go with the simple answer and say fog contributed.
     
    Steve Maller likes this.
  20. Gary McCready

    Gary McCready Active Member

    Joined:
    Oct 26, 2012
    Messages:
    284
    Likes Received:
    42
    Bummer Steve, just saw this. I know if you fly long and often enough you crash....but we all hope it is just something minor. I always feel lucky to just get away with just a few props and an arm. I crashed my old Quad 550 yesterday, trying to get it back, after a voltage drop. Just one landing gear, and a prop, $5. I'm much more careful with the bigger ones....
     
    Steve Maller likes this.

Share This Page