Hello. Just wondering if I could get some advice.
Since day one, my hex has had a problem with uncommanded roll/pitch. I think (its hard to tell with the control issues) I have gotten my rate pids pretty much dialed in. When I find the sweet spot (slightly forward and to the left on the stick) it hovers nicely. The moment I let the stick center the thing rockets off to the right (~10 degrees) and begins a slight drift backwards. I was able to trim out the aft movement, however, even with maximum left trim, the copter still wants to shoot off to the right.
Things I have done:
- Balanced the props with a magnetic prop balancer. I have been using this for a while and have always had good results on my other planes/helis.
- because my hex has tubular arms, I built a fixture that holds my BeastX digital pitch guage and mounts on the prop shaft. Each motor is 0 degrees +/- .2 degrees (with respect to the frame)
- Leveled the frame during the accelerometer calibration to 0 degrees, again using the BeastX pitch guage
- Before each flight, I level the frame with a spirit level and invoke the "preflight sensor calibration" from mission planner. Is this the same as running the "level" command from the CLI? After doing this the AHI is dead level.
- I have tried several vibration control solutions on the APM board. I started with one piece of thick vibe isolating gel under the apm and then moved to four individual (about .5"x.5"x.25") pieces at all four corners with a rubber band providing some downward force on the gel pads. I noticed no difference between these two methods.
- I tried various trim methods (as mentioned above). In most cases I can get rid of the aft movement with a small amount of trim however the uncommanded roll to the right happens no matter how much trim I use. I have gone all the way to 100% left trim. It makes no difference.
- Verified "X" configuration, proper motor rotation, prop placement, etc... The obvious stuff.
The only other thing I can think of is to move the motors around on the frame to see if something changes. Possibly one motor is putting out more/less power than the others? I would think that the APM would adjust for this unless its too extreme. I really don't want to do this as it will require a complete disassembly of my hex.
I will include the log files from my most recent flight as well as an excel file showing telemetry from the Z accelerometer with each motor running individually. I'm thinking that some sort of vibe issue may be causing this but I'm no engineer and this data only kinda makes sense to me.
Sorry for the long post, just wanted to be through. Anything else one would need to troubleshoot this problem I can provide. Thanks for reading and or providing any help!