Dear Collaborator,
We have now entered the 8'Th week of the Au running, the beam is steadily
improving though occasionally there are hiccups. There are two issues for
the plan for the remaining part of the run. The aim of this mail to get
feedback from the collaboration particularly on the overall goal, so the
more routine execution of the data-taking can be carried out with this clear
goal in mind.
One issue is the strategic overall goal: what physics do we want to obtain
in the 2001 run? The other is the practical issue of how we obtain this
goal, i.e. how many settings and how much time at each of them, and
scheduling of runs? At the end of the mailing are added information about
the present status of the machine. Please comment on the plans put forward
here.
The primary goal is to ensure that we measure hadron
spectra (with good statistics) at a number of rapidities. Initial
measurements should be around the expected mean pt of pion/proton, then
extending these measurements to higher pt. With the current and expected
luminosity it is not possible to achieve continuous rapidity distributions,
though we can do pt spectra for a few selected rapidities (y=0,1, and at
large y, 2-3).
A plan that would cover this reasonable well is to collect data for
Theta = 2.3, 3, 4, 8, 12 and 20 deg with the 2.3-8 in 2-3 field settings
(and two polarities)
and the 12 and 20, 1 setting (2 polarity) with the forward spectrometer, and
at 90, 35,40,45
with the MRS with two field settings (and two polarities) with minimum goal
of collecting > 2,000
protons for the 0-10% centrality bin in a ~30 % region around the reference
momentum.
A more extended goal is to get good statistics for less central collisions,
which easily
requires 5-8 times the running time per setting. The strategy should be to
ensure the
coverage and statistics for central collisions before moving on to longer
run needed
for less central collisions.
The actual running up to this point has been to a) ensure we get data at
higher rapidities
both to overlap last years settings, b) and to explore a wider y and pt
range by taken
data at 8 and 3 degrees and c) collect high statistics data for mainly the
90 deg,
but also 45 degree.
For the day-to day data taking other considerations come into play as a)
detector conditions,
b) machine background, need for calibration runs etc, specific requests for
zero-field,
voltage settings. A practical consideration is that the 2.3 deg running, as
well as low to
intermediate momentum runs where the full FS are in use should be done with
C1 out
of acceptance. Such runs should be done at a time where background
(hopefully)
are less since the T5, H2 and RICH has to work well for such runs. Access is
required
before and after such switch over. Tactically we also need to know about the
performance
of detectors, quality of runs, good information about actual 'tracks'
particle, collected.
This brings us to the second issue. As discussed at collaboration meeting a
long time ago,
and brought up by Jens Jorgen recently with a specific proposal we need to
analyze data
as we go along. We are also at a point in time where this has become
feasible. A lot of
effort has taken place to develop code to perform calibrations, overall
tracking, and
particle identification using TOF and Cherenkov systems. Truly these will
have to be further
developed, but are in such shape that it should be used in a combined way
for several purposes
a) Get feedback to experiment on quality and quantity, and to help detect
problems with detector components, performance.
b) Further check quality of data, code and calibrations
c) Create output data, (root files, trees (ntuple) and histograms, log
files) that can be used be the collaboration for further though most likely
preliminary analysis, and code development.
The plan brought forward is to have this routine data processing occur
ongoing with the
experiment, and with the help of people at their home institution having
responsibility for
a given period (in the order of a week at a time) to process the data using
the RCF farms,
and storing the output on the data disks. The necessary programs and scripts
are being
put together by Ian; the system has been checked out. The next step is to
make the
more detailed plan including which other institutions are involved, how the
output is
made available to all etc, and to process in a systematic way.
This kind of coordinated effort is also a model for how we can later perform
'final'
calibration and analysis passes on the data. I have talked to Ian and
and he is preparing a detailed implementation plan for data reduction and
calibration.
Just as the efforts from all collaborators are needed to take the data,
such an efforts
is needed for the analysis with the hopefully much more data to come in the
coming weeks.
// status ..
The machine is now improving a lot, particular in regard to reproducibility.
RHIC may
have reached a limit on ions per bunch with a max current around 25-30
10**9 ions,
so the focus is on the beta* of 2 acc, and increasing to 120(112) bunches.
Presently
RHIC fairly routinely gets to 300 ZDC/sec and can go to ~700 with beta*=2.
This
implies that in the operating conditions with 1h re-fill time one can
achieve ~300-400K /hour,
and with 50% availability for beam 5M/day, which will be sufficient for the
program.
With such rates it is necessary to impose more stringent triggers. My
suggestion is to
implement the vertex trigger, rather than the centrality trigger. The
vertex trigger will
sample in a region +-20 cm with high efficiency, rather than attempting to
utilize a
wide vertex which can results in quite varying normalizations from run to
run.
best regards
Flemming
------------------------------------------------------
Flemming Videbaek
Physics Department
Brookhaven National Laboratory
tlf: 631-344-4106
fax 631-344-1334
e-mail: videbaek@bnl.gov
This archive was generated by hypermail 2b30 : Sun Oct 07 2001 - 19:03:07 EDT