Hoping someone might be able to help me with this log. I was testing my quad in some wind (15-20mph). It flew a mapping mission well but when it came to land it was descending fine then got into some nasty oscillations about 10 feet before touching down. I put it back in loiter flew forward into a grassy field and if flew fine going forward but had a bit of a bumpy landing as I set it down.
Some of the build components are :
6S
Tmotor MN5008 340KV with 18" Tmotor props
TMotor Air 40Amp ESC
750 sized quad
Cube Orange on a mini carrier board
Mauch power
I have run Auto tunes before. The desired vs actual pitch and roll can be seen here and could be a bit better but They seemed to go kinda nuts at the end right before I went back to loiter
Hi Owen,
I think motor3 going to minimum sometimes is bit of an issue. Carefully check the frame and see if you can spot any motor mounts slightly twisted away from perfectly level. Also check the weight/center of gravity is quite well balanced.
Let’s make the changes below and go from there.
You should be able to leave ARMING_CHECK,1 unless there is some particular component giving you trouble.
INS_HNTC2_ENABLE,0 not needed
INS_HNTCH_FREQ,50
INS_HNTCH_BW,20
INS_HNTCH_REF,0.16
INS_HNTCH_FM_RAT,0.70
Then probably do a test flight and provide that log - or if you think flight is good move on to doing Autotune one axis per session. I can see you’ve probably done Autotune before, but these changes should allow it to produce a better result.
Thanks for the quick response I will check those things The motor mounts unless bent cannot twist as the arms are not round but I’ll check for bent or uneven mounts and also the balance. For motor 3, is it possible that it is a wiring/connector issue or is what you are seeing the FC is powering motor 3 down intentionally?
I’ll make those parameter changes and get back to you.
I made the suggested changes (except the land on critical battery vs rtl. Want to look at that more.)
It was not ideal for auto tune as there was a 8-10mph breeze.
Flight 1- autotune roll. Initially it failed after 2 min or so (failing to level). I changed to alt hold vs loiter then did auto tune again and it worked.
Landed and cycled power
Flight 2 - auto tune pitch. It never finished after 9 minutes I put it back to alt-hold or loiter and landed and cycled power
Flight 3 - no auto tune. I hovered in loiter for 30 seconds. Changed to alt hold and flew some figure 8s. Changed back to loiter and landed
I still see dips in motor 3 and 1 (RCOU CH1 an CH3) The 2 front motors.
The frame is rectangular s square so I expect the pitch pids to be close but not identical to roll. I can try copying roll values to pitch and fly again.
The Roll attitude control is definitely much better than pitch now, and quite acceptable.
The dips in output for motor 3 seem not as bad, but it probably wasnt a demanding flight.
I would say increase this: ATC_ANG_PIT_P,12.5
to bring Pitch more inline with Roll
I’m just hesitating in making more small changes, because this probably fixes the issue you identified, and the copter should be flying better than it was. Another pitch autotune might come good on a nice wind-free day.
The BATT_FS_CRT_ACT,1 is quite important because once the copter reaches the Critical battery level it’s likely an RTL will not be possible, it will need to land to save itself.
Before it reaches Critical level, the Low level will have already triggered an RTL, and triggering RTL again has no effect (plus it wont save the copter, as indicated above).
There’s virtually no possibility of the Critical battery level being triggered without first triggering Low battery level. If that happened you’ve likely had some complete battery failure and the copter will be in a lot of trouble. If the battery level does quickly move from Low to Critical then clearly there is a problem and the best course of action is Land immediately.
I’m not saying this is happening in your case, but just for reference:
If battery failsafes trigger sooner than expected, then:
a) the battery pack is ageing or not being charged correctly
b) the voltage sensor calibration needs checking
c) change the timeout value or select the sag-compensated voltage source, which assumes current is accurate
Never change the tried-and-true voltage levels from the values in MissionPlanner Initial Parameters (or in my spreadsheet)
I would just change that one I specified, the other PIDs are all very close.
Ignore the suggested ranges in MissionPlanner - that is old news and these days (with filters and all) higher PIDs are easily achievable.
looking at the desired vs actual pitch I am not sure they are any better than last time. and the Vibrations in Z seem bit high but maybe that is in the acceptable range?
Vibrations are good, quite acceptable.
Attitude control is good, I’m not sure if it’s better than earlier but it’s not bad. At this stage it depends on how you need it to be - is it worth spending more time on, or is it good enough now?
Sorry been out off town and not been able to fly more. Last flights went well. The main thing I am concerned about is still the hover in wind (15-20mph) and when it descends vertically in its own prop wash. Forward movement seemed good. Once back I’ll do more flights and see how it goes. Thanks for all you help.
