Yawing Moment Cn and Rolling Moment Cl calculation: flawed (negative when supposed to be positive)?

[Klausisi-arch]

Calculating a Sideslip polar in Flow5 (latest version), the Yawing Moment Cn seems to show a negative value, when it should be positive and vice versa. Tested it with the Jibe Project by Andre. I get the correct value in XLFR5.

Same thing with rolling moment when setting one flap up and other down: Moment point in wrong direction.

Can you replicate may result?

[Flying-Geek]

Hi,
have a look at this discussion from a while ago as well as the frames & axes and stability documentation.

A positive sideslip can be seen as the nose of the aircraft pointing to the left relative to the oncoming air. This can also be seen in the plane view when a non-zero sideslip operating point is selected. The axis around which the moment is calculated is from the pseudo wind coordinate system, which has its Z axis pointing upwards, the same way as it is done in AVL.
With that, one can see that a positive sideslip (nose pointing to the left) produces a restoring moment that wants to yaw the aircraft to the right. Since the axis around which the moment is defined is pointing up, said moment is negative (right hand rule).

First Correction [THIS IS WRONG]:
I initially misunderstood the combination of the AVL & flow5 documentation. If the definition does indeed follow AVLs standard stability axes (and at that the common stability axes convention) then the yawing moment coefficient should indeed be positive for positive angles of sideslip.

Since the initial explanation is the correct one and in line with the latest version, I have restored the explanation.

Stefan

[Klausisi-arch]

Dear Stefan,
thanks for the swift clarification!

I now understand the convention. I am still confused however, that the direction of the yaw moment displayed in the planeview does not point into the restoring direction as shown in attached screenshot.

[Flying-Geek]

For that I don't have a good explanation, I also usually don't look at those. Could possibly be a bug. André, could you comment on that? @techwinder

[techwinder]

The short answer is that the moments are given in stability axes to be consistent with AVL, with x pointing foward, y right, and z down. See the Type 5 comparisons on this page.
I understand that it may be confusing because flow5 only shows the geometric axes , with x pointing backwards, y right and z upwards.
Even so AVL offers different options see the 'Output' section of the documentation:

 The moments can be defined in four possible ways:

             Body axes     Stability axes
          ---------------  --------------
Geometric|   X   Y   Z        x   y   z
         |
Standard |  -X   Y  -Z       -x   y  -z


Rates    |   p   q   r        p'  q'  r'
Moments  |   Cl  Cm  Cn       Cl' Cm' Cn'  

so maybe flow5 should do the same. Or use the standard aero dynamic conventions:

• The roll, such that the starboard wing goes down is > 0
• The yaw, such that the nose goes to starboard is >0
• The pitching moment nose up is > 0

I'll give it some thought.
André

[techwinder]

There has been some back and forth in flow5 to determine in which frame of reference the moments and stability derivatives should be calculated.

The project file linked on the jibe2 doc page was built with v7.01. However the sign conventions were changed in v7.24 which explains the discrepancy that you report. Since then the moments are given in wind axes and the stability derivatives in stability axes.

Also up to v7.56, the wind axes in flow5 were analogue to AVL's "pseudo wind-axes" with the sideslip ignored. This choice was made for the benchmark comparisons to be valid. Starting in v7.57, the moments output in flow5 will be "true" wind axes. The difference however should be minimal for small sideslip angles.

There is no plan to change that in the future. I will update the documentation and the project files to clarify the conventions and to make everything consistent.