Strange Twitching- Need help with logs

[Goldie]

I'm hoping someone can look at my logs and see what might be going wrong.

I got a parts box solo. It was completely disassembled. I repaired the broken connector and reassembled.

I've taken it out 2 days now, and something isn't quite right.

I've calibrated everything.

At certain points during flight, it will oscillate, or kind of twitch. It will be hovering and then just suddenly twitch and one side will dip down for a SPLIT second, then it recovers.

Today I was landing and just a couple of feet from the ground and it did it, but didn't have time to recover and rolled over upon landing.


I can tear into it and see if something is loose, but first I'd hoped someone could see if they see something in my logs.



Here's a link to the files. They're in the folder separately and I also zipped them if you want to do it that way.

Please let me know if you can look and see anything.


ownCloud

 

[andrew.baker.142]

I'd guess it is a failing esc or bad connection to/from one. Not sure if the log file will show if it were. I had the same issue on a 250 quad. Would randomly drop a corner just for an instant. Replaced all 4 esc's and problem solved.


Sent from my iPad using Tapatalk

 

[Goldie]

Here's a pic of the graph.

How do I tell which corner dips?

 

[pyrate]

OK I am just passing on knowledge shared with me from much smarter people. I learned that looking at RCOU will tell you if you have a motor or esc struggling to maintain. in your log motor 3 is higher than the others but not significantly so,. Notice the mean values for each. that could just have been wind or weight, the errors seem more significant.


This method is now in all my post flight log analysis for my DIY as I have 8 motors

 

[pyrate]

might try a stick calibration, this is from the quick analyze the throttle was having an issue
Log File C:\Users\Admin\AppData\Local\Temp\tmpACE1.tmp.log
Size (kb) 21085.7177734375
No of lines 292994
Duration 0:09:33
Vehicletype ArduCopter
Firmware Version solo-1.3.1
Firmware Hash 7e9206cc
Hardware Type
Free Mem 0
Skipped Lines 0
Test: Autotune = NA -
Test: Brownout = GOOD -
Test: Compass = GOOD - mag_field interference within limits (5.53%)
Max mag field length (584.11) > recommended (550.00)
Test: Dupe Log Data = GOOD -
Test: Empty = FAIL - Empty log? Throttle never above 20%
Test: Event/Failsafe = GOOD -
Test: GPS = GOOD -
Test: IMU Mismatch = GOOD - (Mismatch: 0.35, WARN: 0.75, FAIL: 1.50)
Test: Motor Balance = UNKNOWN -
Test: Parameters = GOOD -
Test: PM = NA -
Test: Pitch/Roll = NA -
Test: Thrust = NA -
Test: VCC = GOOD -

 

[pious_greek]

Thanks for the heads up on RCou, looks valuable... I was looking for a way to analyze the new t-motors, ones seems to spin with slightly more friction than the others...

RCOUT (pwm output to individual RC outputs):
RC1, RC2, etc : pwm command sent from flight controller to the esc/motor/RC output​

@Goldie , I think you also have to consider which way it rolled and pitched compared to how the solo desired itself to roll and pitch. It's the difference that tells you what solo failed to do.

For example, if roll and pitch difference are both negative in the area where it dipped, then it dipped left and forward, so solo's front left motor may be a concern. It's attitude information is below:

ATT (attitude information):
RollIn The pilot’s desired roll angle in centi-degrees (roll left is negative, right is positive)
Roll The vehicle’s actual roll in centi-degrees (roll left is negative, right is positive)
PitchIn The pilot’s desired pitch angle in centi-degrees (pitch forward is negative, pitch back is positive)
Pitch The vehicle’s actual pitch angle in centi-degrees (pitch forward is negative, pitch back is positive)
YawIn Tthe pilot’s desired yaw rate as a number from -4500 ~ 4500 (not in deg/sec, clockwise is positive)
Yaw The vehicles actual heading in centi-degrees with 0 = north
NavYaw The desired heading in centi-degrees​

 

[Goldie]

I grabbed it and just spun the motors by hand.


The back right #4 motor is definitely harder to spin then the others. Not a lot, but it feels like there's something not right. I may have to dig into it.