Present: Burin, Franz, Mac, Nicolaj, Rich, Robert
Agenda:
1. alignment
2. DC efficiency
3. online monitoring
item 1:
Inside a single sector it is often difficult to find out whether
a high Chi^2 of the track fitting is due to misalignment or
an inexact doca function.
Therefore, Nicolaj determined doca functions for each sector.
They differ esp. for Sectors 1&6 wrt. the other sectors
(T_max is about 50 nsec shorter).
In order to compare Nicolaj's alignment studies with those by Robert
there will be two approches:
(1) take the geometry corrected by survey data as starting point
as done by Robert; i.e. change Nicolaj's (design) geometry setting
to the actual survey data and do global fitting.
(2) set the geometry to Nicolaj's numbers and do global fitting.
(3) In order to get a data sample that is not biased by field
corrections for the minitorus (was in during the B=0 electron run
in February) we request for a B=0 photon run at empty target.
In order to collect a data sample of ca. 100,000 'good' tracks
that run would last for ca. 1 hour.
An appropriate variable to define the quality of the fitting is
the left-right residual defined by the difference between the fitted
track and the drift circle (or isochrone) with positive sign
if the track passes the cell outside the drift circle and negative
sign if the fitted track is inside the drift circle.
The doca functions obtained via SSLT show little deviations for
different superlayers (in the order of ca. 0.5-1.0 mm).
item 2:
Although the 'time vs. wire#' plots (for each layer&wire in the
system) suffer from low statistics, they indicate 2 phenomena:
(1) for a small amount of channels the t0's (extracted via pulser runs)
are off by 10-15 nsec,
(2) the t0 is smeared out for a few groups of neighbouring channels
belonging to single HV groups. Missing field wire HV causes a less
precise drift behavior that should be taken into account in
acceptance and efficiency calculations:
for these channels at least the doca has to be smeared by a few mm;
it's also possible that these channels have partly to be knocked out
if we'll find an evidence that their efficiency is much worse
than for other channels.
At least the wire-status map should contain a flag for these wires
besides 'hot wires' and 'dead wires'.
item 3:
We should make some effort to improve the online monitoring, esp.
to help shift takers to find and classify DC pathologies.
For this purpose the DC statistics bank and a rough analysis of
existing histograms in the event_monitor program should be taken under
consideration.
===============================================================
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
---------------------------------------------------------------