Request: Better Yaw-iTerm management for winged quads and / or wing aircraft

[ResinSmoker]

Is your feature request related to a problem? Please describe.

Quad mode works perfectly, no issues or problems what-so-ever. However when in winged-mode, the Yaw experiences "Windup" and Bounces-Back hard during turns. Thus far the only way to minimize this effect is to fly smoothly and with as little Yaw input as manageable.

Note: With tail sitter, Winged-Quads Yaw and Roll effectively swap as the craft transitions to winged flight.

Describe the solution you'd like

I've two thoughts here...

1st solution: A feature where the Yaw iTerm behaves and tracks its set point normally while no Yaw RC-Stick input is given. Then once RC-stick input is provided, disable or scale back Yaw-iTerm (proportional to the stick input) to allow the craft to turn without suffering Windup / Bounce-Back.

2nd solution: Have the Yaw track its setpoint normally until a large deviation takes place. Then simply change the setpoint to the currently held orientation. Effectively this would allow the Yaw-iTerm to function normally yet discard its setpoint once the deviation becomes too great or is held too long.

Describe alternatives you've considered

I've tried reducing Yaw iTerm, adding iTerm-relax and Yaw specific relax without having any success managing the Windup seen. Increasing the amount of relax means the bounces are smaller but much more frequent. Decreasing / removing relax means the bounces are less frequent but much stronger. Furthermore, disabling / reducing Yaw iTerm can result in the craft Rolling wildly due to the presence of wind. Thus as such disabling Yaw iTerm is not an option considering the outcome.

Video , Images and other relevant material

YouTube videos of flight #2 and #3 First FPV flight: https://youtu.be/_OxOEXYaWHk Flight after Wing / Motor swap-out: https://youtu.be/kODrXkHS4UU

Betaflight 4.2.9 installed, using UAV-Tech filtering / motor presets for 3" Whoop.

This Quad flies using an 18650 LiPo and has lasted almost 17 minutes hovering around or 14 minutes during winged flight. Keep in mind that this craft is a WIP and its performance should improve the more I work on it. My ultimate goal is to pull the throttle down to 40% during forward flight and achieve flight times in excess of 20 minutes. Obviously the biggest challenge currently is not being able to turn effectively without the Windup rolling the copter wildly.

[mituritsyn]

It is hard to investigate without log file. Do you have I-term relax enabled on RPY? After enabling, try to adjust cutoff

[ResinSmoker]

@mituritsyn

Problem with most builds using 1S (18650) is that there is really nothing available that has Black Box. Installing a 3rd party option such as OpenLagger is problematic due to the limited number of ports available for what amounts to a Whoop FC. (one for the FrySky RX and one for TX control)

Edit: I wish mfr's would jut standardize BB for every controller made. Would simplify things a bunch.

To answer your question: iTerm relax has been previously tested as you've suggested, with the cutoff set at a range of multiple values. At best all this does is change the frequency of the windup and its intensity. Granted, lower cutoff values perform better but the copter is only barely manageable in turns.

In other news: I'm trying new wings for both upper and lower mounts. Just testing to see if the dissimilar wing sizes could be creating a situation where the wings are fighting against the copters attempts to Yaw (Roll in fwd flight). If this is true then it means that the wings are self-stabilizing Yaw (Roll) as the larger wing set orients itself above the copters CG. (See previously attached image to get an idea of what I mean with regard to the wing sizes) Beyond all of this I'll be resetting BF to the PID defaults as this appeared to work the ok'ish for the initial flights.

[ResinSmoker]

Today I flew with two set of wings.... Long and Stubby. (They came out really nice too)

Note: iterm_windup = 50 and iterm_relax = 10 (PRY)

Here is what the long (170 mm tip to tip, 48 mm wide) looks like... Video link https://youtu.be/1njnwBWP6jY Seams like the wing length (tip to tip) was too much and causing the Yaw (Roll in winged mode) to fight against the wings. Once the copter Pitches / Rolls to prepare for a turn the windup releases.

Here is the Stubby (Approx. 98mm tip to tip, 48 mm wide)... Video link https://youtu.be/K7VVIE-FNQQ Overall this flight was very smooth and even tried a few tight turns towards the end of the flight.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs within a week.

[github-actions[bot]]

Issue closed automatically as inactive.