Hi Michael,
I have a test ring with a horizontal emittance of 16nm-rad. If I add a vertical chicane (i.e. several CSBEND elements with TILT=pi/2) and recalculate the emittance, I get a value of 41nm-rad (using both the ex0 from the twiss file, and ibsEmittance) - note that the optics in the horizontal dipoles is identical in both cases. Looking at the components of I5 and I2 I see that the vertical bends are providing an increased I2 term (as expected), but that the I5 term also increases (there is no seperation of planes for the I5 term, so is it the horizontal I5, or some combination of horizontal and vertical?). The emittance then blows up by the expected change in I5/I2 ratio. However, unless I am mistaken (and I frequently am), the horizontal emittance should not include the contribution from I5 in the vertical CSBEND elements. In fact, shouldn't the horizontal emittance drop, due to the increased I2? Obviously the vertical emittance will blow up, but I'm not so bothered about that at the moment...
Pete
Horizontal Emittance in a ring with vertical bends
Moderators: cyao, michael_borland
-
- Posts: 1933
- Joined: 19 May 2008, 09:33
- Location: Argonne National Laboratory
- Contact:
Re: Horizontal Emittance in a ring with vertical bends
Pete,
Sorry for the late reply.
The &twiss_output command doesn't handle radiation integrals from vertical bends correctly. You should get correct results using &moments_output, which has a self-consistent 6D treatment. You'll need to have an RF cavity with proper voltage, phase, and frequency to get valid results.
--Michael
Sorry for the late reply.
The &twiss_output command doesn't handle radiation integrals from vertical bends correctly. You should get correct results using &moments_output, which has a self-consistent 6D treatment. You'll need to have an RF cavity with proper voltage, phase, and frequency to get valid results.
--Michael