I finally got my Hawk Sky out for its first full autopilot flights to work on tuning the PIDs and I have things mostly dialed in (over 2hrs of flight time on 3 batteries, all successful), but I noticed that the throttle pulses (high, low, high, low, high, low...) constantly when in auto mode. I think this is because my airspeed sensor is not setup correctly (or its bad?), as the airspeed jumps all over the place in rapid succession.
Obviously I will work to correct the airspeed sensor, but is there a way to set some sort of delay or averaging in the throttle changes by chance, so its not constantly switching speeds for the entire flight? This can't be good for the motor, ESC, and battery life.
In the mean time I've basically had to pin the min/cruise/max throttles all around the same values, which is far from ideal. But even with a perfectly functioning airspeed sensor, a gust of wind could easily change the speed enough to cause a throttle change that only lasts a couple seconds.
Also I was trying to find a list that describes all the PID settings themselves (and what they do), but have been unable to find one so far. I know there is the Getting Started guide on PID tuning, but that only touches on a very small subset of the actual available values.
Thanks.
Replies
Did you make sure that your Energy/Alt PID wasn't turned up too high? This will cause the throttle to pulse in order to keep perfect altitude. If you turn the gain down the pulse will get much less noticeable, at the expense of maintaining altitude + or - a couple of meters instead of feet. You could have a combination of interference and gain set too high also.
So I moved my pitot tubes to the side of the fuselage about an inch away and separated the active and static tubes by about 1cm and setup a bench test.
With the Xbee enabled and transmitting (to test for interference), I held a small house fan up to the tubes and didn't see the "0" readings like I did out in the field, though I did see a few here and there throughout the test. It did seem quite sensitive to the direction that I held the fan though, but once I got it just right, things seemed "fairly" stable, in the 3-7m/s range.
This was all without the battery plugged into the plane (ESC). As soon as I plugged that in, immediately the airspeed sensor jumped to 4-6m/s and hovered around that (still no 0 readings though). So it appears that the majority of the interference is caused by my ESC as Joscht originally suggested. :(
Mike i have the same problem. Please check airspeed log (i assume you use airspeed sensor enabled),,, I have problems with my airspeed sensor.. values are jumping from normal to 0. This might be the cause of your problem. There might be some induction from ESC but i have tried dozens of solutions but no luck for now.