Roll control issues

[xrayfishx]

Hey y'all.

Today I created my first successful variable-sweep plane. Now, it flies really well, being able to pull many Gs at any speed, but it has one major issue. At a dynamic pressure of about 37kPa the roll controls get reversed. Basically, I can see the ailerons moving like they're supposed to, but the result of that movement is the plane rolling the wrong way. I press E, the left aileron goes down, right aileron goes up... and the plane rolls left. Well, in the first version of this plane I fixed the issue by disabling said ailerons at that pressure using dynamic deflection and enabling the roll control on flaps, which I set up as reverse ailerons, basically. That fixed the issue.

However, I didn't like some other things about that particular plane so I decided to overhaul it a bit. The current version is better in every way, but the roll issue became a lot more severe. After a lot of fiddling around I still haven't managed to get it working as supposed to. I have reached a point where it won't roll left when I press E or vice versa, but at 32kPa to 41kPa it has very little roll control, with the worst point being at around 36kPa.

So what could possibly cause the roll control to become reversed like that?

The relevant mods I'm using are: FAR, AJE, Infernal Robotics, B9 Proc Wings.

Here are some pics. Click for a larger version.

Low sweep mode.

High sweep mode.

I understand that some of my text might be difficult to understand. I'm bad at explaining stuff. If you have problems understanding what my problem is, tell me and I'll see if I can explain it better.

[ForScience6686]

Does the sweep action alter the com? I thought that has caused some issues for some people but I would think it would cause some horrible flight characteristics as well.

[RoboRay]

Roll reversal (as well as yaw and pitch reversal) is something that can occur naturally with certain wing shapes at higher mach numbers. I'm not an expert on aerodynamics (just a pilot), but I wouldn't be at all surprised if FAR models it fairly accurately.

It's also perfectly normal for control surfaces to lose a lot of their effectiveness in the transonic region. You want to get established at the desired course and attitude, then punch through as fast as possible to regain effective control when supersonic.

Edited September 19, 2015 by RoboRay

[xrayfishx]

Does the sweep action alter the com? I thought that has caused some issues for some people but I would think it would cause some horrible flight characteristics as well.

The sweep alters the CoM, but it has this problem regardless of sweep (although it's worse in the low-sweep mode).

Roll reversal (as well as yaw and pitch reversal) is something that can occur naturally with certain wing shapes at higher mach numbers. I'm not an expert on aerodynamics (just a pilot), but I wouldn't be at all surprised if FAR models it fairly accurately.

It's also perfectly normal for control surfaces to lose a lot of their effectiveness in the transonic region. You want to get established at the desired course and attitude, then punch through as fast as possible to regain effective control when supersonic.

Well, this isn't neither high mach speeds nor transsonic. This is a bit before transsonic, it's mostly fine when I'm already going above 300m/s. But just below that is where the problem occurs.

My current hypothesis is that it's due to how easily the IR parts bend. When the control surface is supposed to push the wing down, it only pushes the rear part of the wing down while the front part of the wing stays where it is and as a result the wing produces more upwards lift, not less. And the opposite would be happening with the other wing. I figured I could test that, and if I'm right, fix it using quantum struts, but after installing them they don't appear under any category so I'm looking for another way to reinforce these wings.