Can wrong declination cause resistance to yaw?

I’ve been trying to tune a small medium size quad (electronics are a bit cramped) using ch6 while flying fpv. Setting declination min max to 0 - 3.0. I’m located in Rio de Janeiro and my map declination says it should be -22° 25’ WEST. I’m starting with the knob set at mid point and adjust the declination more negative. What I’m seeing is when I fly large circles to the left it makes very smooth turns but when I try right hand turns and I get a type of roll wobble and the yaw is much less responsive. Sometimes I have to force the yaw all the way to the right and it still doesn’t turn. I’m using banking right at the same time. When I make left turns this doesn’t happen but the opposite happens, it will sometimes even continue to spin to the left even though I stop giving it left yaw.

Does this indicate that I should be tuning more negative on the declination? When I move the slider down to it’s max I read a value of -0.511731 so I know it’s working. I also see the difference in resistance to turning right when I change the value between 0 and -0.511731 but I can’t get rid of the resistance. When I just fly forward and zig zag left and right, I don’t feel the resistance but when making large circles I do.

Should I try increasing the min max range more than 0 - 3.0?

Here is a video I did that sort of shows the problem. All left turns are smooth but with a bit of overshoot or overtendancy to turn left. Right turns are just the opposite and I have to use a lot more yaw stick to do what I can do easily on a left turn.

youtube.com/watch?v=aVi4Vwka … ata_player

Set your declination parameter to -0.391244641 in the full parameter list (or set it to zero and turn on autodec), leave it that way and look for other causes of your problem.

Setting it to something like 3.0 will cause your yaw to be 180 degrees off, which could crash your copter even in stabilize

Ardupilot uses acceleration as a reference to get attitude - acceleration due to gravity always points down, but there’s a problem: Since our accelerometers measure both gravity and maneuver accelerations, we instead compare to the GPS, which lets us cancel out our maneuver accelerations.

Now, the GPS measurements are in the earth frame (north-east-down) while the accelerometers are in the body frame. We rotate the accelerometers into the earth frame in order to align them with the GPS. If yaw is off by 180 degrees, we double the aforementioned problem instead of canceling it out. This could be construed as a bad thing when flying a circle (which causes continuous maneuver accelerations!)

Thanks. I understand it better now. I’m surprised I didn’t have a crash. I will set it to auto dec and see what value it comes up with and then fix it there. I thought there was a problem or risk using auto-dec but maybe that was in a previous version. My declination is -22° 22’ W.

The compass offsets and compass-motor calibration are a more likely cause of compass issues.

Just to close this post. It ended up that the problem was a twisted arm causing unnecessary yaw resistance.