[Jeuke26]

Hello all,

Went out flying yesterday with my new LS-210 frame and all motors and props are balanced. Wen flying streaky circles at a decent speeds the quad attitude is near perfect ( flaying rate mode only ). But as soon as I apply throttle to speed up or to correct for obstacle the quad starts to wobble high frequency on its roll axis only.

I suppose this is PID related, when hovering everything is fine without excessive vibrations not in smooth flying.

Might this be a PIDthrust scaling issue to be set?

Thanks.

Philip

[NicholasDavid]

Roll P gain maybe too high. But yes you can set the TPA to just work on one, two, or all three axis. Or P,I, or D. I would try -.10% reduction on roll P in the TPA graph at the right(high) end and zero at the middle and left (low throttle) end of the graph

I had the high throttle shakes on one of my copters but couldn't decide if it was roll or pitch so I just did -10% and that smoothed out full throttle punches. I should probably tune it some, but I was in a hurry and anxious to fly it

[Jeuke26]

Roll P gain maybe too high. But yes you can set the TPA to just work on one, two, or all three axis. Or P,I, or D. I would try -.10% reduction on roll P in the TPA graph at the right(high) end and zero at the middle and left (low throttle) end of the graph

I had the high throttle shakes on one of my copters but couldn't decide if it was roll or pitch so I just did -10% and that smoothed out full throttle punches. I should probably tune it some, but I was in a hurry and anxious to fly it

Thanks a lot, will give it a try this weekend. I didn't know TPA existed before I was looking around in Librepilot 
I always thought that the PID loop is a dynamical system that automatically alters with speed, inertia, etc. Seems strange to me that a quad can fly perfect in a Biver or slow circle and exactly the same in high speed sharp cornering and throttle punches with the exact same PID the whole time. Or, is it possible? Just meticulous tuning without TPA?

[NicholasDavid]

I like to think of tuning as calibrating all your motors, esc, frame, weight balance ect to work together with software(PID). A "perfectly" tuned copter (if it exist) is always in a range, if that makes sense.  Sometimes your components are a pretty good match for each other. Some time they are not. And TPA is a way of widening that range.  Some copters need TPA more than others. It's all dependent on your hardware components, flying style....

[Jeuke26]

I like to think of tuning as calibrating all your motors, esc, frame, weight balance ect to work together with software(PID). A "perfectly" tuned copter (if it exist) is always in a range, if that makes sense.  Sometimes your components are a pretty good match for each other. Some time they are not. And TPA is a way of widening that range.  Some copters need TPA more than others. It's all dependent on your hardware components, flying style....

Ontried usine TPA function and dient make a difference. Finally i broke a prop and went from 3bladed 5050props to my spare 5045BN two blades props and the issue was gone completely.

Must admit that enabling the TPA made a difference as well, especially on the P gain for the Roll axis. So thanks for the suggestion.

[Mateusz]

If your pitch is too high you may increase the risk of vortex ring state and flight becomes wobbly and unstable. I think ecalc how bad would be it warns about it when pitch is ridiculous too wide compared to prop length.

Sent from my Nexus 5X using Tapatalk

[Jeuke26]

If your pitch is too high you may increase the risk of vortex ring state and flight becomes wobbly and unstable. I think ecalc how bad would be it warns about it when pitch is ridiculous too wide compared to prop length.

Sent from my Nexus 5X using Tapatalk

Suppose you are correct as going from 5050 tris to 5045 two cured the high speed flutter.

Still looking for good yaw PID guidelines. I'm close but far from perfect.