I am having problems tuning the loiter mode in a big quad: it fly very well in stabilize mode (after a hard PID tuning work) and quite well in AltHold, but it’s unstable in Loiter (it can hold position, but easily starts oscillating, crashing if you don’t change to stabilize mode).
My setup:
Quad +
1100 mm motor to motor. DIY carbon fiber arms frame.
T-Motor U8 135 kv.
T-Motor 30A opto pro ESCs.
28" T-Motor Props
6s batteries
Pixhawk
APM:Copter v 3.1.5
Tranis + X8R
I tried to tune Loiter PIDs by diferent ways, but without sucess.
has anyone experience tuning loiter mode with big props (28") and slow motors (135kv, 6s)?
I am more curious about the mounting arrangements for the pixhawk and where its mounted in relation to and how far the separation from power wiring , by same token whose gps/external mag arrangement used? 3DR? some other brand?
if 3.1.5 then only one of the mags used(not sure which)?
if 3.20-rc? then BOTH mags are used and aligned graphically…
almost sounds like the oscillation is a form of toilet-bowling
and yes logs will help greatly…
hzl
I am planning on building/buying the same size rig so my interest is NOT academic…
50.bin is my last test with these supposed improvements:
3DR gps/mag on a 9 cm pole.
Compass recalibratión.
compassmot.
ESCs recalibration
ESCs timing to 26.25º
Stabilize PIDs: P=0.1, I=0.05, D=0.01 (nice results in stabilize mode; increasing D was the key to kill oscilations).
Loiter PIDs: Varying searching for a solution.
ANGLE_RATE_MAX: Varying searching for a solution (180000, 4500, 3000).
Low but variable wind.
The props balance was not the best due to some previous damages, but it seems that the vibrations stay within the acceptable.
11.bin was the only flight that loiter worked well in the very begining of my tests (in later flights with the same parameters but with some wind the results were catastrophic -> Crash):
No compassmot.
Original Stabilize PID values.
Original Loiter PID values.
Absolutly no wind and very gentle sticks movements.
For a large copter like that, I think I would suggest:
ANGLE_RATE_MAX lower - currently 180 deg/sec. I’d set it to 60 deg/sec for such a large copter.
Setting up a tether for tuning a copter of this size is strongly recommended. The copter should be free to rotate about the axis you are tuning, but not free to crash.
Decrease STB_PIT_P to 3
Increase RATE_RLL_D in 10% increments until you start to get rapid oscillations. Back it off until you don’t ever get them.
Increase RATE_RLL_P in 10% increments until you start to get less rapid oscillations.
Set RATE_RLL_I to half of RATE_RLL_P
Repeat last 3 steps for pitch
HLD_LAT_P 0.5
HLD_LON_P 0.5
LOITER_LAT_P 0.7
LOITER_LON_P 0.7
LOITER_LAT_I 0.25
LOITER_LAT_I 0.25
May need to lower THR_ACCEL_P and THR_ACCEL_I. May need to lower THR_ACCEL_I in proportion to THR_ACCEL_D (make them equal)
May want to use 3.2 release candidate so that you can use the EKF. It is probably much safer. Same suggestions apply, but ANGLE_RATE_MAX has changed name to ATC_RATE_RP_MAX
I think I have a good solution for RATE_RLL and RATE_PIT PIDs, obtained following a protocol similar to that you propose.
The rest suggestion seems interesting, I will try it.
May need to lower THR_ACCEL_P and THR_ACCEL_I. May need to lower THR_ACCEL_I in proportion to THR_ACCEL_D (make them equal)[/quote]
I meant THR_ACCEL_P here. So THR_ACCEL_I = THR_ACCEL_P.
The ratio depends partly on how quickly the copter responds to throttle requests. Since it is a large copter and the props have a lot of inertia, I would suggest backing off the I term in relation to the P term.
Hello guys…
I’m trying to figure out my issues with tuning my U8 135 kv too.
My setup is:
Quadcopter X4
U8 series 135kv + 29" tmotors proppelers
ESC Xrotor Hobbywing 50A professional
PixHawk 3dr as FC
Battery 6S
What PID values have I put to improve my stability flight?
It is very unstable.
I don’t know what else I can do for solve it.
please, if you guys knowing something, help me.
Thanks in advance!
