S800 Major Crash - The spin of death

[junanata]

Yesterday we did a test flight with our recently purchased Zenmuse Z15 and our helicopter failed in flight. We just took off and leveled out around 100 ft. While in flight our helicopter started spinning out of control and dropped out of the sky for no apparent reason and crashed damaging the helicopter and the Zenmuse Z15. We just recently upgraded the software to 5.24 last week and also upgraded the IMU to version 2. Our battery was fresh; 25.2 volt when we took the reading before fight. That was the first lift off with that battery. Judging from the video below does anybody have any idea what cause the crash? <strong>

 

[ProPilotWannaBe]

Yesterday we did a test flight with our recently purchased Zenmuse Z15 and our helicopter failed in flight. We just took off and leveled out around 100 ft. While in flight our helicopter started spinning out of control and dropped out of the sky for no apparent reason and crashed damaging the helicopter and the Zenmuse Z15. We just recently upgraded the software to 5.24 last week and also upgraded the IMU to version 2. Our battery was fresh; 25.2 volt when we took the reading before fight. That was the first lift off with that battery. Judging from the video below does anybody have any idea what cause the crash?

So sorry about the accident. I know how heart breaking they can be.
Can you tell us some other stuff? I have several popcorn questions that might help those in the forum diagnose. Like what flight mode you were in? How did the pilot try to respond? Did he try switching modes? Apply throttle? Roll/pitch correct? Do you know if any wire cam undone? Like the GPS cable coming out (which happened to me in flight), or the IMU cable? Was this your first maiden flight after the installing the Zenmuse or the IMU v2?

 

[junanata]

So sorry about the accident. I know how heart breaking they can be.
Can you tell us some other stuff? I have several popcorn questions that might help those in the forum diagnose. Like what flight mode you were in? How did the pilot try to respond? Did he try switching modes? Apply throttle? Roll/pitch correct? Do you know if any wire cam undone? Like the GPS cable coming out (which happened to me in flight), or the IMU cable? Was this your first maiden flight after the installing the Zenmuse or the IMU v2?

ProPilot

I was in GPS mode. There was a tree in front of us. As the helicopter flew away to the point where the tree was about to block us, I tried to to slow it down and turn around. That was when it started to spin and went down. My view was blocked by the tree as it went down. It happened so quick I did not get a chance to response. We flew with about 10 times IMU V2 before the accident. This is not the first time we fly with the Zenmuse. We had a few tests before. The GPS came off the GPS mount. I don't if it came off before or after the impact. I did not notice any wire undone.

 

[ProPilotWannaBe]

I was in GPS mode. There was a tree in front of us. As the helicopter flew away to the point where the tree was about to block us, I tried to to slow it down and turn around. That was when it started to spin and went down. My view was blocked by the tree as it went down. It happened so quick I did not get a chance to response. We flew with about 10 times IMU V2 before the accident. This is not the first time we fly with the Zenmuse. We had a few tests before. The GPS came off the GPS mount. I don't if it came off before or after the impact. I did not notice any wire undone.

That's some good information. I'll throw out two theories, and others can chime in with others, and/or refute mine:
1) Your gains are to high, so when you reacted and quickly reversed direction, it caused some FC overreaction due to the tight gains. From there, the tight gains just continued to exasperate the problem by overcorrecting. What are your gains set to?
2) Your GPS/compass was loose, or at least enough that the presumable jerk or quick direction change made it pop off and flop around which then would also cause something similar to what you saw (or didn't see ).
My inclination would be the latter possibility.
It's good to hear that you had some flight time on the new IMU and Zen that gives some confidence that configuration change didn't trigger it.

 

[ChrisViperM]

I agree with ProPilot, although without a decent logging function on all the Flight Controls it's always a lot of guess-work after a crash....


Apart from that, one thing I noticed is that you guys just took off like a rocket straight from the ground. I had it a few times already with that WooChang crap that on the ground everything looked fine, but 20-30 seconds after take-off it lost GPS contact, started to shake, gave me white LED flashes or any other stupid thing. So when flying my S800 + WooChang I get it just a few meters above ground, just enough to get out of the ground-wash and let it hoover for about 30-40 seconds.....during that I switch between Alti-Hold and GPS-Hold a few times and then start the actual flight....and I never do any manouver fast....everything chilled and easy....saved my rig a few times already.



Chris

 

[junanata]

This was our settings: 130 130 140 160120 120. ProPilot what you said makes a lot of sense. I just wish I know the truth. Chris it was an illusion. Things look faster than normal when you are that close to the subject. I had to admit we flew a little bit higher and a littler faster than normal on that flight. Sigh....

 

[eyeball]

Sorry to hear about your incident... but is this the Evo S800 or the older version because if its the older model then your gain seems very low as there normally around the 200 mark?

 

[junanata]

Sorry to hear about your incident... but is this the Evo S800 or the older version because if its the older model then your gain seems very low as there normally around the 200 mark?

It's the older version. I didn't want it to be too sensitive to the stick. I don't know how having it that low contributes to lost of control. I feel ProPilot theory #2 makes more sense because my GPS was kind of loose. However I don't think that it was that loose that it would just pop off like that...I don't know... I will have to check the GPS when I get back to the office Monday.

 

[ProPilotWannaBe]

It's the older version. I didn't want it to be too sensitive to the stick. I don't know how having it that low contributes to lost of control.

I tend to agree that low gains wouldn't cause this type spiraling. I know pilots that like their gains very low for smoothness.

 

[junanata]

Guys. I have a chance to check the GPS. It's not that loose. I don't it pop off and cause the spiraling. If I had to guess I would say maybe be one of the motor was out. I emailed DJI support last week about getting it repaired but the email went unanswered. Has anybody had experience with them?

 

[ProPilotWannaBe]

I emailed DJI support last week about getting it repaired but the email went unanswered. Has anybody had experience with them?

I guess I have never tried to contact DJI directly. For me it would be like trying to get ahold of Toyota in Japan to answer something about my car. I personally always go to my dealer. If somebody is selling the stuff, they should know the stuff, including policy on replacements or stuff like that. My dealer (IntelligentUAS) has been great to work with.

 

[ChrisViperM]

I guess I have never tried to contact DJI directly. For me it would be like trying to get ahold of Toyota in Japan to answer something about my car. I personally always go to my dealer. If somebody is selling the stuff, they should know the stuff, including policy on replacements or stuff like that. My dealer (IntelligentUAS) has been great to work with.

+1....best thing to do. Contacting DJI directly is the same as trying to walk on water. If someone can help, it's the dealer (unfortunately mostly very limited in case of crashes / failures). People running for the cheap and buying directly from Asia or any no-name web-shops are usually really f*cked (sorry for the f-word)


Chris

 

[junanata]

Thanks guys. We got hold of Robert from uavproducts.com (our dealer). We are exchanging emails right now. I hope he can help us. Crossing my fingers.

 

[jhardway]

if you can keep up on the post I would like to see where they keep it going, To me it looks like you may have lost a motor or an ESC, especially seeing it fall like that.

I am not convinced, even though they say the software can configure hexa to fly with a missing motor. unless you shut down the opposite motor directly across from the out motor I do not believe the it will work unless you have a very balanced copter.

In the case of the s800 it is not a balanced machine. Notice in they're video of out motors, the copter is down to a bare bones with no camera mount on it. I think that is because when you add the gimbal you over weight the copter to handle an out motor.

I am not sure but the out motor/esc is just a guess.

Best of luck with it all, If you get to the right DJI person you will not have a problem, I know Robert is in your corner, I have had a few events in the past and each time it has been tough to get it solved, just say with it, you should be OK.

 

[SMP]

That was hard to watch..ouch..sorry guys!

On the brighter side, thats gonna be a helluva addition to your crash reel someday and hey how bout that Zen!!

Hope you guys are back up soon!

 

[junanata]

Thanks for the well wishes guys. jhardway you are right and also just as I suspected; one of the motors failed. We opened it up and found one of the cables going from the motor to the ESC was loose. I don't know if it was loose before or after the crash; but the way it spirals down it looks like one motor was out as seen on the video below. The thing is it shouldn't fall down fast like that should it? I know we have the weight of gimble but if you look at the second half of the DJI video, they also have the copter loaded and it doesn't just drop down like that. It makes me wonder if it is a wookong error causing the copter to drop and the cable got disconnected from the impact; or if it is a result from a combination of failed motor and a wookong failure maintain control.

 

[ChrisViperM]

Did you solder something on the motors/ESC's ?

This solderpoints are thin like paper, and if you don't watch out and soldering is too hot, they come off the circuit board together with the motor wire and the solder.


Chris

 

[Teamflail]

Sorry to hear about your crash. I think we have all experienced that pain. I have been lucky no crashes with gimbal and camera but I am sure my time will come.

ChrisViperM, thanks for the good tip about hovering and switching modes. I will add that to my post take off checklist. I agree about your big maneuver theory and the last time I crashed (2 arms down), was trying to stop the hex quickly in GPS mode. Definitely keeping it very smooth now.

 

[ProPilotWannaBe]

Thanks for the well wishes guys. jhardway you are right and also just as I suspected; one of the motors failed. We opened it up and found one of the cables going from the motor to the ESC was loose. I don't know if it was loose before or after the crash; but the way it spirals down it looks like one motor was out as seen on the video below. The thing is it shouldn't fall down fast like that should it? I know we have the weight of gimble but if you look at the second half of the DJI video, they also have the copter loaded and it doesn't just drop down like that. It makes me wonder if it is a wookong error causing the copter to drop and the cable got disconnected from the impact; or if it is a result from a combination of failed motor and a wookong failure maintain control.

i personally do not think it came undone in the crash. Several months ago I had my S800 crash into my car while it was doing a good 30mph (long story). Though it broke a lot of stuff, none of the solder joints anywhere came undone. Just another data point.

 

[deluge2]

i personally do not think it came undone in the crash. Several months ago I had my S800 crash into my car while it was doing a good 30mph (long story). Though it broke a lot of stuff, none of the solder joints anywhere came undone. Just another data point.

Yikes, there's some serious 'experience' being gained. BTW, which was going 30 mph, the car of the s800? If there were any particular lesson's learned that you care to share, I'm sure we'd be interested. If we learn for both our own and others' misadventures, it makes it harder to come up with new failure modes. Well that the hope anyway.

As regard the OP, I agree it seems likely that the loose cable probably was the cause (or at least the primary cause), rather than the result, of the crash. junanata, can you provide any more detail (or a pic) of the 'loose cable'? That might help distinguish fatigue-related breakage (away from solder joints) from solder point failure for various possible reasons. Were the connections 'factory stock' or were there post factory changes made?

Steve