1. 100k Km21, Km22 and Kp21 UMC raw data are now avaliable in
/sitka/scr22/e949/umc
2. The pass2 ntuples are also avaliable in
/sitka/scr5/e949/umc
3. New features in the UMC data
a) New beam file for 2002 is used.
b) Km21, Km22 and Kp21 triggers are corrected to
accommodate the change in E949.
c) Since the UTC center is shifted by 1 cm to match
the center of T counters, the beam
stop position in UMC
should be switched back in E949. The Z offset in I-counter remains
the same as in E787.
4. How to run an UMC pass2 job
a) For pions
/sitka/home/e949/umc/run/pass2_umc_pion_simple.csh
>pass2_umc_pion_simple.csh
/scr22/e949/umc/kp21.dat
b) For muons
/sitka/home/e949/umc/run/pass2_umc_muon_simple.csh
>pass2_umc_muon_simple.csh
/scr22/e949/umc/km21.dat
5. Some comparisons between E949 and UMC data
a) Kp21 P/R/E (discrepancy
in E only) , K stop (discrepancy
in xy), stopping layer
and energy, timings
b) Km21 P/R/E (some
shift) , K stop (discrepancy
in xy), stopping layer
and energy, timings
c) Km22 P/R/E (some
shift) , K stop (discrepancy
in xy), Energy in each
layer , at the stopping
layer, timings
6. Shift in the range and the energy distributions
Because we can not measure the full muon
range and energy, a correct beam file is necessary. However, the kaon stopping
position distributions in the x-y plane of the target
have somewhat difference between data and UMC in both km21
and km22
monitor samples, which indicate that data have more kaons stopping close to
the edge of the target,
giving a downward shift in data as a consequence.
To confirm this, we put two cuts on the dip angle |cos3d|
< 0.1 and
|sqrt(tgx^2+tgy^2)-3.5| <
0.5 cm, a study of
km22 sample shows the magnitude of the shift is reduced. A further
study
is to check the energy
deposit in each layer between data and UMC, which the black dots represent
data, the open triangles
resprsent UMC and the open squares are for the ratio
of energy of data over that of UMC for each RS layer. All of them
are very close to 1.0, indicating a good muon simulation.
For some reasons, the 2002 RS energy calibration,
the muon energy is well measured, but the pion energy is about
2 MeV below what we expected. These do not agree
with what we observed in 1995-1998 data, where the pion energy
is well measured, but the muon energy is about 2
MeV above the expected value (see page 12 in 1998 pnn1 analysis
for more details). Since pions also suffer nuclear
interaction in addition to the minimum ionization, we expect the
pion energy should be somewhat smaller than the
muon even if the calibration is perfect.
7. Discrepancy of kaon stop position in the target x-y plane
According to Takeshi's message, the E949's km21
trigger accepts (17ct+18ct+19ct), while the E787's km21 trigger accepts
19ct+20ct+21, and Peter's KOFIA code analysing
km21 for beamfile did not have the last RS-layer requirement.
It was turned out that about 18% of the events in
the previous beamfile was those whose stopping was 17 or 18.
By choosing the km21 events whose RS stopping-layer
is 19, a new beam file was created. A test has been done,
the result
(where the open triangles for those with new beam file, while the histogram
are for those with old beam file)
shows that the change actually goes to the
opposite direction.
9. The thickness of T-counter
In E949, the T-counters were all replaced.
The thickness of T-counter is measured to be 0.7cm in stead of 0.635cm
used in E787. The offline stopping layer
is used for the test, the result
(where histograms are for the km21 data, open
circles are for the old UMC and the black
dots are for the new UMC) shows some improvement.
10. Energy loss in the target and IC
In UMC, we observed a clear correlation between
the P/R/E and
the target and IC range. However, the correlations don't
show up in the Kp2 momentum
and range
for E949 data except the energy.
We check the 1998 data and get the same
results for P/R/E. The
correlation between the energy and the target and IC range is understandable,
because the
saturated target and IC energy is added to
the unsaturated range stack energy.
11. The BVL TDC information
A BLCT bank is created to keep the BVL TDC information
so that the kofia code bl_tdcunp.F can easily fetch the
timings information. To do so, the UMC code tapwrt.F
is updated.
12. The RD and BVL timing precision
In real data, the RD and BVL timing precision is
0.5ns/count, but the previous UMC gives a 1ns/count precision.
Therefore, the UMC code tapwrt.F
is updated.
13. Work to be done
a) Transmission efficiency in each RS counter and
resolution effects are not included in UMC
b) Target and IC simultation
...