This is a summary of the recent status of DC calibration. No alignment
issue is discussed here.
1. Cable delay calibration
The content in the current DC_TDLY.map is the result of pulser run 8088.
The data were taken in Dec. 97 and analyzed using a special recsis user
package. The calibration constants are applied back to Sep. 97 after taking
into account a few TDC cable changes.
The date analysis yielded no information for 18% of the total channels
(sense wires). Since we assigned a same constant to every eight channels
(half of those in one cable), there are only 11% which have no information
and were assigned the values of their nearest neighboring channels.
** Fine tuning of the cable delay constants is necessary to further improve
TBT. Accumulated TDC spectra per signal cable can be used for the purpose. **
2. Absolute t0 determination
The absolute t0's can be extracted from the peak positions of derivative
TDC spectra which include corrections from start time, TOF and signal
propagation time along the sense wires. The current t0's are from run 8774.
** The t0's have to be determined on an effectively per run base (at least
per running period). Since the t0's are correlated with the above mentioned
corrections and SC calibration constants, they should be reevaluated
when necessary. **
3. ** Signal propagation speed adjustment
The structure of the mean value of the residual distribution vs phi,
the azimuthal angle, indicates the speed currently used (1/2 speed of
light) is not optimized. This has to be adjusted for better TBT. **
4. Single layer efficiency
The single layer efficiency was studied using single super layer tracking.
The HV scan results (run 6103-6110) showed the nominal settings are
reasonable.
5. XvsT
A new xvst conversion has been implemented in recsis. The xvst correlation
was studied using single super layer tracking (straight track fit for
Region 1 and curvature fit for Region 2 and 3). The parameters were
optimized for run 8774.
The new xvst features track local angle correction and magnetic field
dependence (Region 2 only). The field strength range the study covered
is between 0.2 to 1.0T. ** The extrapolation to zero field has to be tested
and information at higher magnetic field is desirable. **
The absolute t0's and small doca behavior of the xvst conversion are
determined by electron induced tracks. A "timewalk" correction compensates
the different response of particles with different beta relative to electron.
The correction affects most the hadrons of large and small beta.
6. Maximum drift time tmax
The maximum drift time the most important parameter in the XvsT conversion
and should to be determined on a per run base as the t0. ** Before a complete
scanning (mapping) of t0 and tmax for all runs (running periods) is done,
the runs benefit the most from the new xvst are data taken shortly before
and after run 8774. **
7. Single layer resolution
Along with the new xvst tables, a sigma as function of doca is provided
to weight the chi2 for fitting. The average sigma's (single layer resolution)
are 285/310/350 microns for Region 1/2/3, respectively. There are compared
with earlier assumptions of 400/500/700 in recsis. (This is why the chi2
using the new xvst looks worse despite of the improvements of the residual
distribution.)
This information was obtained from single super layer tracking. The numbers
should become better with more accurate drift time calculation.
8. ** Sector dependent problems
The sector dependent problems have not been addressed in the new xvst
setup. The worse resolution in R1S1, R2S1, R3S1 and R3S6 may be because of
bad cable delay constants. Before the situation is improved, sector 2 and 5
might be the only good pair for alignment study. **
9. ** Others you may come up with **