Prompted by Mike's comments, I compared the injection envelopes
and stored and injected beam wall clearance calculations
for bmad_4260_092605_v1 and bmad_4260_min_v1.

The e+/e- separations are impressive not just for the
sigma_seps being different by more than a sigma,
but even more for the qualitative effect of the
near-collision points having moved from the west to the east.
An odd effect is that the parasitic crossings happen at
the peaks of the pretzel EITHER in the east OR the west,
but not both. In general the 092605 lattice does look worse,
as Mike points out. The attached ps files hsep_092605.ps
and hsep_min.ps show the comparison.

The wall clearances for both the stored and injected electron
beams are significantly better in the 092605 lattice - maybe
this is just a different way of saying that the dynamic aperture is
better. The plots showing this are attached: wall_092605.ps
and wall_min1.ps


I also looked at the dependence of the lattice parameters
on positron current for the electron bunches.

By the way, DHR, my original comment that the current dependence
looked very bad was only because I forgot to fix the tunes and they were
changing rapidly with positron current, as has been the case
in all lattices. Fixing the tunes fixed the problem.

It turns out that the increase of beta max with positron current
is much better in 090625 than in min, as can be seen
in the attached plots bmax_092605.ps and bmax_min.ps.

I also looked at the welch-temnykh parameter's dependence on
positron current (bparam_092605.ps and bparam_min.ps).
Both lattices show the familiar effect that
electron bunch 1 suffers more than the other bunches. It is
curious that this is so even though the lattice optimization is
done for bunch 1. I've often wondered if the bad behavior of
the B parameter at high positron currents might be related to
the poor luminosity we see from bunch 1.

-- jim


HTML files of the above ps files can be found here:
hsep_min hsep_092605
wall_min wall_092605
bmax_min bmax_092605
bparam_min bparam_092605

========================================================
James Crittenden                   Tel. (607) 255-9424
Wilson Synchrotron Laboratory      Fax  (607) 255-8062
Cornell University
Ithaca, New York 14853-8001
========================================================


Mike Forster wrote:

Hi folks,

Following the success of Dave's lattice on Monday, particularly with
injection, it is interesting to note that the sigma separation
parameter is much worse in Dave's optics.

DLR's lattice (bmad_4260_092605_v1.lat):

! 9x5 info:  I_bunch:3.26  B_param_i:2.26
! Ring sigma_sep:5.58


The energy scaled lattice, bmad_4260_min_v1.lat, that was used
previous to DLR's (Pretzel was scaled to give the same max
displacement):

! 9x5 info:  I_bunch:3.07  B_param_i:2.15
! Ring sigma_sep:6.86

The b-parameter is slightly better in DLR's.  The e- horizontal beta
at 34E is much smaller in DLR's (32m vs 48m).
This suggests that the separation sigmas don't dominate.
It also calls into question the importance of large beta-x at 34E.
Perhaps all is secondary to the difference in dynamic aperture?

Mike