Message-ID: <47CE6271.9080605@dl.ac.uk>
Date: Wed, 05 Mar 2008 09:05:53 +0000
From: Andy Wolski <a.wolski@dl.ac.uk>
User-Agent: Thunderbird 2.0.0.12 (Windows/20080213)
MIME-Version: 1.0
To: Louis Emery <emery@aps.anl.gov>
CC: "Mark A. Palmer" <map36@cornell.edu>, ilcdr@lbl.gov
Subject: Re: This Week's Damping Ring Discussions at the Sendai GDE Meeting
References: <47CB6B12.5040200@cornell.edu>  <47CD13B2.9000609@cornell.edu> <1204703901.3521.100.camel@localhost.localdomain>
In-Reply-To: <1204703901.3521.100.camel@localhost.localdomain>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Dear Louis,

Thanks for your comments.  You raise some good questions: the dynamic 
aperture in particular is something that we should look at carefully, to 
make sure that we understand the issues, and arrive at an optimum solution.

Regarding the dipole field, this (or, equivalently, the dipole length) 
was used as a variable in the cell design of the DCO lattice to control 
the momentum compaction factor.  But the dependencies are not strong, 
and there is plenty of space in the lattice if it is found that a longer 
magnet would be preferable.  I'm not sure that cost is a real issue.  My 
guess is that to first order, the dipole cost depends on the beam energy 
and total bend angle.  Since all the lattices we considered for the 
baseline have a total bend of 2pi and all operate at 5 GeV, I would 
expect any differences in dipole cost to be small.  The installation 
costs may depend more on the number of dipoles (so, a smaller number of 
dipoles may be cheaper to install, even if each individual dipole is 
heavier), but I'm not sure about this.

I thought the comparison between the OCS-style and DCO-style arc cells 
presented by Aimin was extremely interesting.  I had not viewed a TME 
cell in that way before.  It does raise some interesting upgrade 
possibilities, as you point out.

The layout is also an issue that we need to pursue.  While it is 
necessary at this stage to have a single lattice identified as the 
baseline, it is very important to maintain at least one alternate, so 
that we can understand the influence of things like the layout on cost, 
dynamics, etc.

Best regards,

Andy.


Louis Emery wrote:
> To all,
>
> I agree with the relative scores in the tables.
> I would weight the first table more. But I think
> it would not change the order of the results.
> I'm leaning towards the DCO cell structure because
> of the reduced number of magnets, and I think the
> DA problem is not insurmountable.
>
> I missed the webex session on Tuesday, so I have some
> comments and questions:
>
> I'm not sure how the DCO lattice at 100 degrees/cell 
> ended up with a (much?) smaller
> dynamic aperture than the FODO5 lattice with 108 degree
> per cell. I see that the cell length for DCO is shorter (making the
> dispersion smaller) than the FODO5 FODO
> cell, perhaps to partially compensate the larger I5 created by the
> shorter and stronger dipole. I think the dynamic acceptance of FODO
> cells with constant circumference should scale roughly as the cell
> length squared. So a shorter dipole magnet resulted in a smaller dynamic
> aperture.
>
> Andy, was using a 2x field in the dipole for reducing the total dipole
> magnet cost? Could the magnet be longer without disrupting
> the mechanical layout much?
>
> To rephrase a point by Aimin, if maintaining a 90 degree phase advance
> is important for dynamic aperture, then an "upgrade" of momentum
> compaction from 1.7e-4 to a value 25% lower can be done with the
> conversion of the DCO cell to a TME-cell (keeping 90 degree
> cell) by the insertion of one quad. This is, of course, at a cost of
> 192 magnets.
>
> How does the lower symmetry of the DCO lattice affect the
> dynamic aperture? Could we have the superconducting
> stuff together in two groups at a 90 degree azimuth relative to 
> the injection and extraction (like FODO4 but with doglegs added to
> reduce heat load)? 
>
> Thanks.
> Louis Emery
>
>
>