Tuning stops at lower HS

[waferthin]

I setup another B2 on my TDF as a replacement for a Spirit.
For whatever reason (linear servo drive? gremlins? small fuzzy animals trapped in the cyclic, who knows) - I was unable to get the Spirit to fly nicely. The most obvious negative characteristic being that it'd keep on travelling in a direction of rotation after a stick movement. It made it very, very weird to fly.

e.g: jab elevator backwards, the heli will pitch up, stop and then .....
slowly pitch up a few degrees more (like it's moving through treacle) in the same direction of travel. *after* I've stopped moving the sticks.

Anyway, I didn't write this to complain about the other controller (I still run Spirit on another heli and it flies great), not at all - simply to provide background.

With defaults the B2 is running already really nice for me. I use 90 agility, 716 rails @ 1250 rpm... just slow 3d stuff. I like that kind of style.

I still want the heli to just 'stop' when I return my sticks to a neutral position.

However I'm seeing a very small amount of the same behaviour that I saw on the spirit (continued movement after stopping stick movement, sometimes).

I've tried upping the cyclic overall gain from (I think it started at 48) 58 and up to 70 with no real discernible difference.

To tune the stops on cyclic, Is cyclic "overall gain" (P gain) the place I should be starting here? I was a little nervous to go to 70. No real reason apart from I don't want my heli to fall out the sky!

Anyway, I thought I'd ask what parameters I should start with to try to tune this out. Tail is holding wonderful with reasonably low gains and only minor adjustments to I gain. As mentioned above cyclic is pretty good, I'm just after that last 5% of locked feel.

As I sit and write this I wonder if in fact D is more applicable to damp the stops...)

Any tips appreciated! Thank you.

[Xrayted]

Originally Posted by waferthin

I setup another B2 on my TDF as a replacement for a Spirit.
For whatever reason (linear servo drive? gremlins? small fuzzy animals trapped in the cyclic, who knows) - I was unable to get the Spirit to fly nicely. The most obvious negative characteristic being that it'd keep on travelling in a direction of rotation after a stick movement. It made it very, very weird to fly.

e.g: jab elevator backwards, the heli will pitch up, stop and then .....
slowly pitch up a few degrees more (like it's moving through treacle) in the same direction of travel. *after* I've stopped moving the sticks.

Anyway, I didn't write this to complain about the other controller (I still run Spirit on another heli and it flies great), not at all - simply to provide background.

With defaults the B2 is running already really nice for me. I use 90 agility, 716 rails @ 1250 rpm... just slow 3d stuff. I like that kind of style.

I still want the heli to just 'stop' when I return my sticks to a neutral position.

However I'm seeing a very small amount of the same behaviour that I saw on the spirit (continued movement after stopping stick movement, sometimes).

I've tried upping the cyclic overall gain from (I think it started at 48) 58 and up to 70 with no real discernible difference.

To tune the stops on cyclic, Is cyclic "overall gain" (P gain) the place I should be starting here? I was a little nervous to go to 70. No real reason apart from I don't want my heli to fall out the sky!

Anyway, I thought I'd ask what parameters I should start with to try to tune this out. Tail is holding wonderful with reasonably low gains and only minor adjustments to I gain. As mentioned above cyclic is pretty good, I'm just after that last 5% of locked feel.

As I sit and write this I wonder if in fact D is more applicable to damp the stops...)

Any tips appreciated! Thank you.

Hi Neil. I saw you replied over on you other thread about this, but wanted to add a screen image from the Brain software that may help you out on which way to start tuning the D gain. You will need to use the Advanced menu to access the individual gain components and placing your cursor over any of them will give you the pop up bubbles describing what effect they will each have on the model.

[trillian]

I had something similar happening with a NextD Total-G FBL unit and it was, if memory serves, something in the stop gains. I think it can also be caused by not having the cyclic pitch range set correctly, like if the FBL needs to calibrate to 8 degrees and you actually had 10 degrees of actual pitch, if you see what I mean.

I know this is not specific to the Brain but as I have experienced something similar - just throwing this out in case it creates an 'aha' moment :-)

[gwright]

Need to adjust the Derivative Gain in the PID loop for the axis you need to tune. the proportional gain (primary, or "overall" gain is the way we normally think of it) isn't going to do what you want very well. Derivative Gain is to fine tune the stops, assuming the proportional is somewhere in the right ballpark.

[waferthin]

Perfect. I'll head that direction (D gain) when I'm next at the field.