The OPAL Discussion Forum

["Dr. Christopher Mayes" <cmayes AT stanford.edu>]

I Christof, what I’m talking about here is laying out the machine. For example, to have two vertical bends in the same direction:

MAD

B1: SBEND, tilt = pi/2

B2: SBEND, tilt = pi/2

OPAL

B1: SBEND, phi= pi/2

B2: SBEND, phi = 0

That is, the tilt convention in OPAL tilts the orientation of every bend afterwards, whereas the MAD convention does not. In MAD, this has the advantage of being able to easily define vertical bends while keeping subsequent elements ‘upright’. 

Chris

On Jul 27, 2017, at 1:59 AM, Christof Metzger-Kraus <christof.j.kraus AT gmail.com> wrote:

Hi Chris, 

the difference between MAD and OPAL is that we can have overlapping fields. Therefore we cannot rotate the bunch only at the beginning and exit of a bunch (or any other element). 

Christof

Am 27.07.2017 12:53 vorm. schrieb "Dr. Christopher Mayes" <cmayes AT stanford.edu>:

Hi OPAL developers,

In MAD and Bmad, the convention of a tilted bend is to rotate the coordinates at the beginning, and the exit, of a horizontal bend. The result is that a purely vertical bend will keep the ‘y’ axis pointing in the same plane. This is also consistent with how other elements are tracked: a tilted quad does not tilt the pipe downstream. 

In OPAL-T, it seems that the rotation (set by ‘psi’) is at the beginning of the bend, but not undone. Is it possible to make this more in line with MAD and Bmad? 

Thanks,

Chris

________________________________
Dr. Christopher Mayes
+1 607 222 5922
cmayes AT stanford.edu