Hello,
Recently I designed a simple quad using a Pixhawk and 3DR Compass/GPS unit to control it. Here is the setup:
A30-42S-UAV 850Kv motors, 12x4.5 MR APC props, PixHawk Controller, 3DR Compass/GPS unit, MinimOSD and Fatshark telemetry, 3S 5000mAh main battery and 3S 800mAh telemetry battery. RTF weight = 1570g (3.45lb), distance between left/right motors=360mm (14.2in), distance between front/rear motors =560mm (22in). Throttle at hover = 54%. Gains adjusted by Autotune. Software: MissionPlanner
This quad works quite well in my opinion (I really do not have a lot of experience with quads) except for a specific situation which generally happens when the quad slows down from speed and also sometimes when descending. I have noticed that when this happens I am in Loiter or altitude hold mode and while sometimes the quad has been able to recover on its own while in those modes, most of the times it has not. The only way that seems to allow me to recover most of the times is by switching to stabilize mode.
Anyway, what happens is that as the quad slows down from speed, or it starts descent with some forward speed, it simply starts to severely wobble in pitch/roll, to the point that it is not capable of maintaining altitude. I have attached some images of the log file for review (the log file is 9MB) which shows the perfect example of this around minute 9. At this point I had started a descent while moving forward slightly. If you check Pitch vs DesPitch and Roll vs DesRoll you can see what I am talking about. At the same time the current draw from the motors is spiking, which seems to indicate that the quad is trying to recover from the oscillation. Going into Stabilize mode and letting it lose a bit of altitude allowed me to recover this time.
Pitch vs Despitch and Roll vs DesRoll shows a very clear divergence between them. Is this some sort of delayed motor response issue?
If this info helps, when moving forward the quad seems to be flying smoother when flying in Stabilize mode as opposed to Loiter mode. Altitude hold mode is sort of in between the two. Again, gains adjusted by Autotune
So, what is happening?
Thanks
Sportjet
Replies
I forgot to add the log Auto Analysisa;
Log File C:/Program Files (x86)/Mission Planner/logs/QUADROTOR/1/9-8-15. No Wing/2015-09-08 11-38-10.log
Size (kb) 9740.5517578125
No of lines 132628
Duration 0:10:13
Vehicletype ArduCopter
Firmware Version V3.2.1
Firmware Hash 36b405fb
Hardware Type
Free Mem 0
Skipped Lines 0
Test: Autotune = UNKNOWN - No ATUN log data
Test: Balance/Twist = GOOD -
Test: Brownout = GOOD -
Test: Compass = GOOD - mag_field interference within limits (15.46%)
Test: Dupe Log Data = GOOD -
Test: Empty = GOOD -
Test: Event/Failsafe = GOOD -
Test: GPS = WARN - Min satellites: 5, Max HDop: 4.9
Test: IMU Mismatch = GOOD - (Mismatch: 0.27, WARN: 0.75, FAIL: 1.50)
Test: Parameters = GOOD -
Test: PM = FAIL - 55 slow loop lines found, max 10.70% on line 16985
Test: Pitch/Roll = GOOD -
Test: Thrust = GOOD -
Test: VCC = GOOD -