Loiter issue, no mag data, bad motor balance

Log Analysis -- deprecated Copter Logs -- deprecated
1

Greetings from France to all you ArduCopter wizards!

On my first flight with a replaced Pixhawk2 I experienced a couple issues that I’d love to get solutions for.

Summary:

  • Tarot 650 + PH2 + M8N + GoPro Gimbal
  • Short flight in steady, normal wind
  • Primary objective: Check if flyable & GPS working (lots of resoldered cables)
  • Secondary objective: PID Autotune

Course of Flight:

  • Behaved fairly sluggishly, but was controlable
  • Hover throttle too high, reset MOT_THST_HOVER from .35 to .5 which improved it
  • Switched to AltHold, good results
  • Switched to Loiter, copter immediately took off to the right until I switched back to AltHold
  • Hood came off, hit propeller, hit Rx antenna, pushed Rx antenna into prop, one Rx antenna cut off
  • Tried to land, touched down, forgot I was in AltHold. Thrust didn’t decrease quickly enough, wind blew it into some high grass where it auto-disarmed
  • Had to call it a day, didn’t want to risk flying with one Rx antenna. Therefore didn’t AutoTune.

LogAnalyzer:

Size (kb) 4423.6962890625
No of lines 50632
Duration 0:02:42
Vehicletype ArduCopter
Firmware Version V3.5.0-rc5
Firmware Hash f070a8d0
Hardware Type 
Free Mem 0
Skipped Lines 0
Test: Autotune = NA - 
Test: Brownout = GOOD - 
Test: Compass = GOOD - No MAG data, unable to test mag_field interference

Test: Dupe Log Data = GOOD - 
Test: Empty = GOOD - 
Test: Event/Failsafe = FAIL - ERR found: CRASH 
Test: GPS = GOOD - 
Test: IMU Mismatch = GOOD - (Mismatch: 0.37, WARN: 0.75, FAIL: 1.50)
Test: Motor Balance = FAIL - Motor channel averages = [1787, 1749, 1542, 1516]
Average motor output = 1648
Difference between min and max motor averages = 271
Test: Parameters = GOOD - 
Test: PM = NA - 
Test: Pitch/Roll = NA - 
Test: Thrust = NA - 
Test: VCC = UNKNOWN - No CURR log data

Issues:

  • Loiter mode lets quad take off to the right
  • Plugs for hood don’t work well (I’ll try some bobby pins next time)
  • CoG seems off due to gimbal (Motor Balance off even though ESCs are calibrated)
  • There seems to be a mag issue. I have the compass inside the M8N activated, also compass 2 (internal on PH2)
  • Motor Balance: Could that be due to a bad CoG from the gimbal?

I uploaded the .bin and two .tlogs (I’m not sure which one is the correct one)

I would very much appreciate if someone would take a quick look and tell me if there are additional issues or what the reasons for the ones listed above could be.

Thank you very much in advance!

2017-06-10 17-49-27.bin (1.9 MB)

Pictures of Setup:

IMG_2049.PNG750×908 764 KB
IMG_2047.PNG1334×750 2.16 MB

2

Looking at your log I cannot find any compass data in there.
There is also no battery data.
Do you have these turned on in logging?

Generally if the copter takes off in a GPS mode you would immediately look at the compass.
(After checking Sats and Hdop of course, but yours were OK)
With the ESC’s so close to the PH2.1 it’s internal compass’s are probably very untrustworthy.
You should do a motor/compass calibration?
EKF complained about a variance, so you might even have to turn the internal compass off.

Motors 1 & 2 are working hard while 3 & 4 are not???
Are the motors level?
Is there a twist in the frame or the frame not rigid?
They seem to fighting a yaw tendency so it does not look like wind effect.

Screen Shot 2017-06-13 at 10.49.54 am.png2560×1440 945 KB

Vibrations look good so no real problem there.

Just my quick look.

3

Thanks for the quick analysis!

cannot find any compass data

That seems odd, what could be the reason for that? Is it possible that the compasses were not working at all?

There is also no battery data.

I did not set up the battery monitor yet, so that’s ok. Will do that before the next flight.

With the ESC’s so close to the PH2.1 it’s internal compass’s are probably very untrustworthy.
You should do a motor/compass calibration?

Thanks for pointing that out! I believed that due to the cube (with the compasses) being on top of the mounting plate, it’s far enough from the ESCs. Seems like I was wrong. I like the placement of the ESCs though because it reduces the mess on the top plate. Would a motor/compass calibration help or should I just rely on the external compass on the M8N? I also considered adding another M8N for GPS redundancy, that way I’d have a second external compass.

Motors 1 & 2 are working hard while 3 & 4 are not???
Are the motors level?

That’s the CoG issue that I suspected. I added a gimbal in front and shifted the battery further back. In this flight, I was using a 6Ah battery instead of a 10Ah one. Thus, I suppose the battery weight was not able to outbalance the gimbal weight.

Vibrations look good

That’s great. Seems like the PH2 IMU isolation does a good job.

Thanks a lot for your input!

4

Bump :slight_smile:

(Sorry, I’d prefer not to take off before I know what caused these issues)