Presents: CB, HB, RdM, SF, MG, JJ, YP, FR, TR, TW.
- Report from meetings
- MEB -> MG
- A SpSSS with a MQ is
under-performing and it has probably to be re-measured.
- There was a confusion
in the files reporting the geometry of the D2L107 allocated to slot IP2
left. A swap of V1 with V2 was found. Following this finding, the
already allocated dipole turns out to be the best choice taking into
account the special loss scenario highlighted by SF.
- An important failure
occurred in the heat exchanger pipes of one of the triplets during the
pressure test of sector 7-8. The triplets were isolated and
pressure tests continued successfully. This failure is possibly due to a design feature
and a destructive test of a spare will be performed to reproduce the problem and
find a solution. An issue occurs regarding the access to the
already installed triplets which may need to be un-installed. In this
respect, alignment optics should be reviewed in view of an engineering run without the triplets.
- During a power test one
dipole in sector 7-8
was found with a short-circuit. SF adds that this seems
to be an isolated problem and that it can be repaired in situ.
- SLM -> MG
- There are still
discussions for
the reorganization. The new group
structure will be implemented as of 1st of February 2007 and it will be
effective for the next MAPS exercise.
- Phase advance scans for IR4 -> RdM, TR (pdf
file)
- A problem was observed with
the new matching method while used on IR4 (Jacobian
for low penalty function). It is still not yet fully understood.
- However, several new features were
implemented, such as
- Use of SVD
- Printing the
rank of the Jacobian for each iteration
- Control
of the minimum singular value.
- Such an approach
was used to evaluate the tunability of IR4. Using a linear step
method for the target phase advances the accessible region in the space of
horizontal and vertical phase advances was computed. This region is smaller
than the one originally determined by A. Verdier.
- An algorithm for a random
sweep of the phase advances was implemented (quite lengthy process) and
a larger accessible region was found. The region is smaller on the high-side of
horizontal phase advances,
probably due to new aperture constraints. MG
suggested to make extra runs by forcing the horizontal phase advance in the
high-side.
- Replying to questions by SF
and YP, RdM answered that the previous solutions are always used as
initial conditions for the next matching run and the phase advance step
is of 0.001 radially, with a random distribution of the angle.
- Replying a question by SF,
RdM pointed out that dispersion is not matched to 0 at the IP. SF
suggested to have as a target in the matching a value of around 30%
of the dispersion in the arcs(50cm), which is used for the aperture tolerances.
- It is suggested to plot n1
and the dispersion vs. the horizontal and vertical phase advances to check
that the aperture constraints are fulfilled and the dispersion is small
enough.
- Status of optics database for V6.500 and
polarities for V6.501 -> TR (pdf
file)
- A new polarity convention
for MADX is discussed.
- Previously, the sign
information was in the strength file and not in the sequence file.
- The most complicated case
is the powering of the triplets where power supplies are both in series
and in parallel.
- An issue occurs with MQ and
the adjacent
MQT, which inherits the polarity from the MQ.
- There is need for feed-back
from all interested parties to finalize this.
- SF favors that the strength
of the magnets carry the sign as it was done for all the machines the
last 10 years. He also proposes to have and additional attribute to contain the
polarity information of the element.
- Questioning who is
proposing the change, TR replies that it is Markus and the database
group.
- MG suggest that in
the discussion, the software team and operations (Mike
Lamont) should be included. A dedicated will be organized in January 2007 (Action:
TR).
- DFBXs and aperture model in MADX -> TR
- Some
improvements have added in TWISS
- The apertures and
tolerances can be listed (functionality added thanks to Hans Grote).
- Kmax and Imax for each
element is defined in
sequence file (extracted from the official database), with calib=Kmax/Imax and Kmax as properties of
elements. SF
questions if Kmax is the 2nd derivative of the field. TR will check this (Action:
TR).
- AOB
- Thys found that the
official magnet database contained some errors. An ECR is being prepared to
update the values.
- The strength of cold D3 and D4 was not updated with the change of
optics versions.
- The option of running triplets in IR2 and IR8 at 220 T/m
and 222 T/m, respectively, will be officially recorded in the
ECR. It was highlighted by M. Zerlauth at al. that, due to limitations
imposed by the cables, triplets in IR5 could not be ran at values higher
than 205 T/m (at 7 TeV).
- A code called WISE was written
by P. Hagen (AT-MCS) that gathers
the magnetic data as well the geometry ones and produces a MADX input
representing the machine
as
built. It is stable and complete.
- Thys has worked on the
magnetic and geometric
part and further cross-checks need to be done per magnet.
- For the MBs some features
are corrected by SF, but not serious bugs found.
- For the triplets a lot of
issues were found (wrong assignments of
magnets, wrong direction).
- for the other classes
of magnets the cross-check is in progress (Action: MG, YP).
Colour code:
Black: proposed agenda
Red: outcome
of discussion at the meeting
Last update on 20-12-2006 19:00:00
MG
& YP