floor coordinates with vertical deflection

[Liu_Weihang]

Hi everyone, I found in the calculation of the floor coordinates of the lattice that when there is a vertical deflection at the beginning, even if I supplement the deflection (i.e. Y=0, dY/dZ =0 ), Y will still change if there is a horizontal deflection after that. I don't understand why this is the case, I don't think the horizontal deflection will cause the Y to change. Can anyone explain this behavior?

[michael_borland]

The reason is apparently that you introduce horizontal bending while the line is on a vertical slope. This introduces an effective roll of subsequent magnets.

To understand this, imagine that you have a vertical 90-degree bend followed by a normal 90-degree bend. In the second bend is defined in the vertically-oriented frame, so it results in the beam propagating horizontally again, but to the side. At the same time, the local coordinate system is rotated 90 degrees.

Looking at the theta (yaw), phi (pitch), and psi (roll) angles in the floor coordinate output files is helpful.

--Michael

[Liu_Weihang]

Dear Michael,
Thanks for the clean explanation.