Flight model question

[Wrench 9,848]

requires the 3d model, as the animation of the part must be redone (it's "time" or frames of movement). it's a common fuck up on a lot of the older models. for some reason, the lod viewer moves the meshes to their "correct" location. seen it before.

[+mue 715]

On 9.12.2021 at 11:03 PM, Cliff7600 said:

Yes when the control surface meshes are not the exact same mirrored replicas left and right.
It tends to make the plane turn.

AFAIK that's not correct. Control surface meshes don't have any influence on the FM. You can test this by e.g. linking the mesh node of the right wing to the left aileron. This gives a funny animation but the flight behaviour doesn't change. 

On 9.12.2021 at 11:03 PM, Cliff7600 said:

Most of the time it's the pylons for weapons that are not fully symmetrical and then the aircraft cannot go straight when the pylons are used in the weapon stations entries.

That's correct. Weapon station entries are one of two cases I know of where the 3D model influences the FM (the other case are the landing gear entries).
The weapon mass is attached at the pivot point (not necessarily the mesh) of the node defined by ModelNodeName. If ModelNodeName is missing, then the position defined by AttachmentPosition is used.
In either case only the lateral (left, right and maybe? up, down) coordinates are considered, the longitudinal coordinates (fore, aft) are ignored.

[+Cliff7600 1,148]

I made another test, because one aileron doesn't change its size and position when giving a rolling input in the opposite direction.

I used the left aileron as the left aileron and the right flap as the right aileron. So different position, size, volume, etc... (and flaps as "fake flaps" no 3D entry). Two meshes really asymmetricals.

Indeed you're right : the plane flies perfectly straight without inputs.
So "It tends to make the plane turn" is not correct.

But when testing the rolling response I'm not sure that the rolling to the left was perfectly the same than the rolling to the right (ini entries 100% symmetrical left and right).
I still think that "same mirrored replicas left and right" is the way to go for modeling 3D control surfaces for aircrafts in Strike Fighter.
And I don't really care about defending my opinion when I'm not sure I'm right. It's better to learn something for good than keeping thinking wrong forever.

So you taught me something : it flies straight. Thank you for that 
...but I will still going on asking perfectly symmetrical 3D models ^^

I used FastCargo's F-86A.

Edited December 12, 2021 by Cliff7600