Z axis rotation of wings caused by Dihedral

[Mike Vivaldi]

Hi!

I noticed today that when applying a dihedral with an incidence and the "Rotate all foils to mach dihedral" the axis of rotation is applied for both transforms using the global XYZ axis when such that the result is the airfoils exhibit a composite rotation around the z axis. Basically the axis of rotation for the airfoil should be the normal vector of the airfoil and should first rotate with dihedral then with incidence. I noticed due to this behavior it is impossible to apply incidence to a winglet when using these settings

[Mike Vivaldi]

Here are 2 more clear shots

[Brandon Hicks]

I think you want to use multiple sections and blending to achieve your goal. If you are rotating foils to match dihedral and you want them to meet at the root, you need to start with a small section with no dihedral. 

[Mike Vivaldi]

The issue is still present in 3.39.1

[Brandon Litherland]

I see what you're getting at now.  Rob, if we take a look at one of our blended wing examples from the workshops, trying to put twist into the last section where the winglet is causes the airfoil to rotate "up and down" rather than "in-and-out".  Can you check if it happens on your end as well?

[Francois De Villiers]

Got the same problem when trying to model a paragliding wing with lots of dihedral.  It twists around the y axis on the section vertical z-x plane then dihedral is applied at the end of each section instead dihedral should be applied first and then twist applied around the axis of the section.  With twist applied linearly over 13 sections the twist is total wrong using the wing sections.  The workaround is to remove the twist  in the wing sections and rather apply it  the modify section part with theta.  However using this it apply twist around the leading edge and not around a chosen % point of chord.

Another point i realized is that the profile is rotated to the end of the section dihedral and not the avg with the next section . This also has an influence on the amount of twist.

Regards

Francois

[Francois De Villiers]

Twist applied to paraglider wing using the two wing configurations below and included in the alf.vsp3  Only Cambered Surfaces shown for comparison.  The aim is to apply 4.5 degree positive twist linearly along the span around the 33.3 % chord which is currently impossible even with sections.  

WingGeom_Normal_Twist - Twist not applied correctly with sections.  Problem - Twist is applied before anhedral is applied 

WingGeom_Modify_Twist  - Twist applied in modified section with theta but can only be done around the leading edge and not a specified % of chord.

Future Request

1.  Have the choice to apply twist after dihedral or anhedral is applied

2.  Have the choice to rotate airfoil to the average of the current and next dihedral in the sections and not just the end of the current section dihedral. 

3.  In the modify section provide capability to twist around a specified % of the chord. 

Thanks for the great program and making it available  ! Its amazing  and looking forward to all the upgrades coming.

Regards

Francois

[Mike Vivaldi]

Has this become an acknowledged Issue or does something additional need to be done? I also reported this as an issue on the github.

Thanks,

Mike