Hello,
I just performed an autotune (pitch + roll, successful) on my quad and was wondering what would be a good way to evaluate if the autotune was good or not? It’s generally a bit windy where I am and I had to autotune with a bit of wind as well today, although I tried to keep the vehicle twitch direction at 90 degs to the wind as described in the documentation.
I tried to use the PID Review Tool but could not figure out how to use it correctly. Any help is appreciated.
I’m not stuck anywhere, as the original post says, the autotune completed successfully. I’m just looking for a good way/metrics to evaluate whether the tune is good or not.
Easiest way is to do another flight in Loiter mode. Do some stick inputs working up to full deflection then back to middle. Make the analysis easier and make simple clear inputs, don’t just mash the sticks around. As you’re doing that watch the drone. There should be no oscillations or anything else unusual. Stop if there is.
When you look at the log check the desired pitch/roll vs the actual. They should track and there should be no sign of oscillations (there might be stuff in the log that isn’t visible when you’re watching).
Unfortunatly you can’t tell much from an auto tune flight log.
Not tune related but there are some high Z-axis vibrations. You might want to look into that before you go too far.
There’s also some really strange motor stuff going on here:
But I was fighting for more than a week trying to setup a copter, and basically, I manually obtained PIDs which make it quite flyable, but the PIDs are about the double of what autotune produced, and there are oscillations of about 1º and 7 Hz. If I try to relax the PIDs, the copter becomes unstable during descents.
So Iam basically looking for any help, because I can not understand how to solve the problem…
Not tune related but there are some high Z-axis vibrations. You might want to look into that before you go too far.
Except the landing part, they look like they are within -15 and -5 m/s^2. Aren’t these the acceptable limits?
My props are balanced, everything (motors, battery, peripherals) is fully tightly secured to the frame, the FC is on a vibration isolation mount, and the frame itself is quite stiff. The only thing I’m aware of being out of tune is that my PSC_ACCZ_P is 0.5 when it should be 0.45 (MOT_THST_HOVER) which I will change before the next flight. In your experience, is there anything else I should be looking at?
There’s also some really strange motor stuff going on. What have you got for motors and ESCs?
Ah really? I do not see any impacts of this on any of my flights. Based on my last few flights even with the default gains, the quad is completely stable (some oscillations in the logs but visibly not too bad) at full stick deflections in all directions. The only thing is that I live in a very gusty and windy region and I have to unavoidably fly in gusty open areas. In my opinion the motors have to rapidly change RPMs to compensate for these gusts which is leading to the “strange motor stuff”, do you think this could be the case? I will try to find a large closed indoor space to confirm.
I have EMAX ECO series 1300kv motors and EMAX BLHeli-20A ESCs.
Post a log of a flight in loiter mode (and hopefully better weather) and let’s see what’s going on. I didn’t look at your filtering so make sure you have the logging set to record the data for filters just so all options can be addressed. Those variations in the motors don’t look like wind to me but let’s rule it out.