Quadcopter using NXP Hovergames kit. I’ve updated all the firmware and loaded the latest stable PX4. Radio is fine and I believe I’ve calibrated everything correctly. When I arm and try to lift off, one (or two) motor are slowing significantly relative to the rest and I abort because the vehicle is just going to tilt and flip.
The setup and calibration has been done in QGC (including ESC calibration)
I’ve downloaded the log and loaded into online analyzer. I can see the throttle input and the output to the motor ESCs. Over time, you can see the ESCs are diverging in RPM until I kill the throttle. My guess is that the PX4 thinks the vehicle is drifting and is attempting to correct to keep in position but the drone has stayed essentially stationary for the whole time.
Any help would be appreciated.
I am running into this same issue. I have everything calibrated: all imu sensors and escs. My propellers are on in the correct directions. In my case, I am not using the gps so I cannot engage in takeoff mode. Instead I am taking off in manual/stabilized mode.
What you describe is what I have been observing consistently. Now I have a lot of stuff mounted (offboard computer, etc) on my drone which I cannot describe in detail on this forum, but its important in my case to know it is all there. I took all those extra components off, and after significantly altering PID variables (I raised integral gain for roll rate significantly), I was able to stabilize the drone and take it off. The PX4 gitbook describes instead decreasing K and D gains to achieve takeoff, but that just was not the case for me, which I find quite odd. Still after increasing I gain, I still observed slow oscillations mid-air.
Putting all the components back on now, I cannot takeoff, and have the same problem you have again. In my case I have been doing all sorts of things to shift the weight distribution and provide vibration damping (soft mounting components).
Nonetheless, I still find this whole issue odd from the start. Even though I was able to fly with all those extra components off, I find it odd that the PID control system gains needed to be drastically changed (especially in a way that is ubreccommended). Two things that have been on my mind lately is:
-
That the hovergames kit as I purchased in 2022 does not exactly have the same components as those in earlier kits. This is likely due to economic reasons and I undersatnd, but I wouldn’t be suprised if the motors being used in the kit are very unbalanced compared to those in earlier kits. The motors that came in my kit are all the same. Usually, two are supposed to tighten if you turn the propeller lock clockwise and the other two are supposed to do the same if you turn them counterclockwise. Instead, I have four motors that tighten clockwise. I still am able to spin them in the proper dorections, but the reason I mention this is that it seems to me very possible that quality of these motors are not up to spec with the old ones. I am not sure at this stage.
-
The second thing I have on my mind is whether I may not be taking off fast enough. What I have noticed in one recent trial is that I was able to takeoff in one flight, but after landing the drone and taking off with same PID gains, I was not able to take off. It is possible that there was a mechanical fluctuation between the two flights (propellers coming loose). But I have also been thinking that it is very possible that takeoff time has something to do with this.
If you have solved this problem since you made this post or if anyone else on this forum has some tips, I would be very thankful. I will update if anything becomes clear to me.
