Hi,
Yesterday an octoquad that had been flying correctly for months, was uncontrollable as soon as it took off. I suspect that an ESC is not working, but testing one by one I don’t find any failure.
Reviewing the log I am not able to discover anything wrong except that the attitude was really bad.
Can anyone check this log to find if really a ESC was working bad?
Thanks!
I can not see any one malfunctioning ESC or motor.
What ESCs and motors and props do you have?
There is some instability and additional tuning may help, as confirmed by oscillating motor outputs.
I think the instability could have started because the GPS position was still wandering when you armed and launched, possibly snagging the landing gear and disturbing the copter enough to induce the instability.
EDIT: No, I’m wrong, you seemed to have the copter flying for a while. The instability was worse during descent. Motors were already hitting minimums during flight, so the instability and descent might have been too much to handle.
Still proceed with everything I’ve said below, especially the motor test to determine MOT_SPIN_ARM and MIN after setting MOT_THST_EXPO,0.4
There is a GNSS issue for you to fix too:
Same Node Id 25 set for multiple GPS
and use BRD_BOOT_DELAY,3000 when using CAN GNSS units.
If you have more than two GNSS units, pick the best 2 and remove the third one. Ardupilot can detect three but will only ever use two (at this stage).
Try these to improve the GNSS performance:
GPS_GNSS_MODE,5
GPS_GNSS_MODE2,5
and maybe even
GPS_AUTO_SWITCH,4
Use MissionPlanner motor test to check MOT_SPIN_ARM is as low as it can reliably go. Set MOT_SPIN_MIN a bit higher than ARM, +0.03
Then I would do:
FENCE_ENABLE,1 ← wont let you arm until Home can be set
ATC_THR_MIX_MAN,0.5
INS_ACCEL_FILTER,10
INS_HNTCH_ENABLE,1 ← set this then refresh params to see the rest
INS_HNTCH_MODE,1
INS_HNTCH_REF,0.25
INS_HNTCH_FREQ,60
INS_HNTCH_BW,30
INS_HNTCH_FM_RAT,0.7
INS_LOG_BAT_MASK,1
INS_LOG_BAT_OPT,4
MOT_THST_EXPO,0.4 (depends on motors/ESCs/Props question)
and maybe reduce the yaw demands on motors a bit, at least until there’s verified stability and a good log for use to check
ATC_RAT_YAW_I,0.08
ATC_RAT_YAW_P,0.8
For the next test flight just use Stabilise, AltHold and maybe Loiter if everything is going OK. The Hover thrust may need some time and throttle inputs to relearn if any of the MOT* params were changed.
EDIT:
Vibrations could be a problem but I’ll reserve judgement on that till we see the next log, or even a log from a previous good flight
Thanks Shawn!
About GPS there si one on serial and one on CANBus. I disable CAN-P2_DRIVER, and I don’t get any more messages of Same Node Id 25 set for multiple GPS. I have also modified the rest of the GPS parameters according to your recommendation.
I have done the motor test and adjusted again MOT_SPIN_ARM and MOT_SPIN_MIN. Curiously I have obtained lower values than the first time I did it. Also setting MOT_THST_EXPO, 0.4
The configuration is:
T-Motor U11
FLAME 60A 12S
MF2614 Folding propeller
This is a log of previous flight:
I am going to update the FW and activate the filters and we will try a flight in AltHold.
Use
INS_HNTCH_FREQ,35
INS_HNTCH_BW,18