On December 29th 2013 I bought a X8 ready to fly quad coper from 3D. The first day I was out with it I could run it around a field in stabilize mode with no real problems. So I thought that it was time for me to run an autonomous mission using mission planner. So I did. Well it made it to the first way point then decided to take a nose dive and crash. So I rebuilt the unit and went though all of the parameters and after speaking with tech support at 3D decided to try it again. Well to make a 4 month old story short this thing has let to make ONE successful autonomous mission to this date. I have been reading and studying for months learning about logs and parameters and anything else and have not been able to figure out why this is going on. I do feel that 3D is doing everything that they can to help but at this point I have yet to understand why in the hell this thing will not fly under computer control. I had plans to put an APM into a X8 plane that I have but at this point I am very concerned if I do invest in an APM that it may put my nice big plane into a tree or worse. The fact is I realize that this is my problem and this flight system can do it I just need to find out what it is I am doing wrong. I have attached one log file in the hopes that someone here that is smarter than me can point me in the right direction and get this nice copter into the air. If any more information is need please just ask.
I would be grateful for any help
Thanks
Hey mate,
Sorry to hear about all your crashes. Have you been flying the aircraft in FBWA/FBWB mode before sending it on a AUTO mission. STAB is not a very useful mode to see how the APM is going to fly the aircraft.
Have you got experience in flying normal RC planes? I would have crashed many more times than I originally did if I didn’t have plenty of experience behind me and able to manual save the aircraft.
Jason
Optlock, I had a look, and I can’t find anything glaringly obvious as to the problem. Vibrations look great. The GPS HDOP is good. I do see that the pitch and roll control is not great. Have you tuned it? It’s a bit loose, particularly in pitch. Have you got a camera mounted on the front, or any other reason why the CG isn’t balance?
It’s not obviously as far out of balance as some others I see, but it does seem to have some tendency to drop the nose. But roll control is also not great.
And what battery are you using? it’s showing around 10.7 Volts throughout the flight which is sort of odd.
Looks like you got nailed by a really nasty bug that I found recently. The fix has been backported to 3.1.2 and is being released as 3.1.3.
The bug was in the stability code, which takes roll, pitch, yaw, and throttle, and transforms them into 8 motor pwm values. In a relatively rare edge case (rare enough that it went undetected), with very large roll, pitch, yaw inputs and a very high throttle input, it would output THR_MIN (or close) to all motors.
See figure 1. You can see that the first thing that happens is your CTUN.ThrOut (red line) starts going up. As it goes up, your motor outputs (green line) track it and then suddenly drop to minimum. Right after that happens, the roll error (blue line) diverges from zero.
Now, given the huge magnitude of inputs required to trigger the bug and the fact that your x8 has very low rate gains and very low stb gains, it follows that there must have been some really big disturbance that caused the bug to occur. However, I doubt that the copter will still entirely flip over if you install 3.1.3.
So give that a try, and meanwhile I will look in your log to try to find what the trigger was.
RTL_ALT is 0. This means the copter will either descend into the ground or RTL at its current altitude. I wouldn’t trust it to RTL at its current altitude, I would set the minimum RTL altitude.
Something is very wrong with your compass. Please recalibrate the offsets, move it further away from high-current wires, and/or do compassmot.
I haven’t found a root cause.
Thanks for all of the input. I will have to show my ignorance here and ask what is “FBWA/FBWB”?
Also I will add that when the copter hit the first way point it spun around and the front was pointing directly at me this was something that I did not intend to happen about that time I hit the RTL on the TX and then it started making a hard bank down to my right in my panic “Because it was on a course to hit a house” I might have tried to change the course VIA the transmitter. As far as the balance I am not sure why it would be I have nothing mounted on it at this point and was only flying it to test the performance. I am running a 4200mah Battery.
Why is the MAV distance so far off?
To address the first response. I have three other quads and have been flying RC planes off and on since my early 20’s I am 50 now UGh… 
I am off to fly my other copter today in the mountains of NC and will be back to finish up repairs on the RTF X8 later tonight. I will look very closely to see if I can see anything odd with the balance and change the settings that has been suggested and report back later.
Thanks again for the help!!
Well got the copter back together and have ran a couple of short test flights. I have updated to the latest firmware. But at this point I am just plain chicken to send it on a auto mission. I can fly it manually without a real issue. Just no faith at this point. One issue that I seem to be having is the fact that I can not get simple mode to work on channel seven or eight. I have read the instructions gone though the the prams to see if I am missing anything checked the flight modes ect but nope no simple mode for me.
Again Thanks for the help
Brian
That user was confused, thought you were flying a plane.
[quote=“jschall”]RTL_ALT is 0. This means the copter will either descend into the ground or RTL at its current altitude. I wouldn’t trust it to RTL at its current altitude, I would set the minimum RTL altitude.
Something is very wrong with your compass. Please recalibrate the offsets, move it further away from high-current wires, and/or do compassmot.
I haven’t found a root cause.[/quote]
Jon,
sorry to digress but how did you find that the offsets were wrong.
I found this in his log and everthing is below 150 and above -150. What are good offsets?
PARM, COMPASS_OFS_X, -52.57542
PARM, COMPASS_OFS_Y, 23.24117
PARM, COMPASS_OFS_Z, 113.9590
Moreover I am confused about these logs, they state ArduCopter 3.1.2 but I do see INAV that has been removed in 3.1.1 (github.com/diydrones/ardupilot/ … a17cdbc014)
The log itself is really corrupted, but I’ll try to show you what I mean… The magnitude of the magnetic field should be constant, since we should be measuring only the earth’s magnetic field, which doesn’t change for our purposes. Take a look at fig. 1… The file is really corrupt, but it shows that the field isn’t even close to constant, it varies wildly between about 100 and about 250.
Anyway, the INAV messages may be from an older log? Like I said, the log is corrupted.
I wonder why it is so corrupted? I downloaded it using the right cable. I may still have it in the APM. I have had a few go manual test fights since this crash and it seems to be performing okay. I was thinking about taking it out today and run a basic Auto flight to see if she might go. I shall see if time allows. I am just don’t want to have to rebuild again.
Thanks for all the help.