Present: Mac, Liming, Nicolaj, Robert, Franz
Agenda:
1 -- How to proceed with extraction of t0's for each wire?
2 -- x vs.t functions ready to be used in recsis
3 -- Status of alignment studies
item 1:
T0 is extracted from pulser calibration data corrected by an
overall offset (DC trigger time) and offsets for each region.
When Alex plotted the time distribution of real data for each
wire in the system, he found that the position and the steepness
of the raising edge at t0 varies for several channel groups.
It is unconvenient to extract t0's by looking into lots of time
distributions, esp. when the statistics is not high enough to
define the position of t0.
Therefore a short program will be developped (by Franz) to
accumulate enough statistics.
The 't0' will then be extracted via a fit on the first derivative
where we define 't0' as max.of 1.derivative minus sigma
(to have a way to regard the steepness of the increase near t0).
Another important parameter for the time distribution is t_max
that may be extracted from the x vs. t distribution (flat part near
the border of the cell) or via fitting a Fermi-Dirac (or similar)
function on the upper end of the time distribution.
Because of several uncertainties in both procedures we first
have to test which one is more reliable.
item 2:
SSLT contains x vs.t functions and updated parameters to build
t->x tables, time walk correction tables, and sigma tables,
regarding local angle and field strength (latter only for Region 2).
Up to now the functions are used for all 6 sectors but it might be
convenient to determine the functions separately for each sector.
For the moment the new functions can be called in recsis among other
representations, that will be changed when the functions are
sufficiently tested.
item 3:
Nicolaj presented new results on aligning opposite sectors
(Sectors 1 and 4): Assuming elastic scattering (with the electron
defined by eid0) the proton angle can be predicted and compared
with the direction of the reconstructed 'proton' track.
Without geometry corrections it's not possible to reconstruct
tracks simultaneously in Sectors 1 and 4;
with geometry corrections the deviation from the predicted proton
angle is about 1 mrad.
To check the assumption the angles of the electron track (in Sector 1)
and the second track (in Sector 4) are used to calculate the mass
of the second particle. The resulting mass distribution peaks at
935 +/- 60 MeV -- consistent with the assumption of elastic scattering.
Robert showed new results of the simultaneous fit in Region 2:
the residuals in R2AX look reasonable when using additional software
offsets (wrt. survey data), except for Sector 6 (whole chamber)
and Sector 3 (theta>50 deg.). This is consistent with Nicolaj's
results showing the greatest shifts in Sectors 3 and 6 wrt. ideal
geometry.
===============================================================
Franz J. Klein, Thomas Jefferson National Accelerator Facility,
12000 Jefferson Avenue MS 12H, Newport News, VA 23606
Email: fklein@jlab.org, Tel: (757)269-5879, Fax: (757)269-5800
---------------------------------------------------------------