Hi everyone,

I am currently working with a rosette transfer setup in the CATIA Composite Design app and encountering an issue with fibre orientation.

Context

I am applying a rosette using a Cartesian transfer on a half-model of a rim, which is later used for simulation. For the Layup definition I use the Grid Approach.

Problem Description

In the section highlighted below, I observe an unexpected behaviour:

The fibre orientation suddenly changes within this region. Additionally, this behaviour is not constant. It varies along the same section.

Observations

• This is the only region where the issue occurs.
• It is also the only section that is oriented 90° relative to the rosette.
• All other areas are either curved or oriented at angles different from 90°, and they behave as expected.

What I Have Tried

• Using different rosette orientations (placing them under a angle)
• Switching transfer types:
  • Cartesian
  • 0° guided by curve

These changes either had no effect or worsened the result.

Impact

The resulting fibre orientations are incorrect and lead to physically implausible simulation results.

Question

Could this issue be related to:

• A limitation of the Cartesian transfer near perpendicular regions?
• Or an issue with how the surface is interpreted locally?

I would appreciate any guidance on how to:

• Stabilize the fibre orientation in this region, or
• Properly define the rosette/transfer method for such geometry

Thanks in advance for your support!

Best regards,
Tim