[Lowq] Minutes of the eg6 meeting on March 2

[Hovanes Egiyan]

  Minutes of the March 2 meeting


       From Lowqwiki

Present: Rafayel, Carlos, Bayram, Stepan, Hovanes, Yohann

On the phone: Eric, Raphael

Notes by Hovanes


[edit 
<http://clasweb.jlab.org/rungroups/lowq/wiki/index.php?title=Minutes_of_the_March_2_meeting&action=edit&section=1>]


    DC calibrations

    Last week Bayram did 3 iteration with two kinds of fits: x-vs-t and
    residual fits.

        * After residual fits the numbers for the residual on the
          summary sheet looks better, but the residual versus DOCA plot
          shows some variations.
        * When x-vs-t fir is performed, the variations of the residuals
          vs DOCA do not disappear but they get larger. As a
          consequence, the residual distributions become even worse.
        * This needs to be discussed with Mac. It is not clear if doing
          the residual fit provides reasonably good x-vs-t function.

    Stepan also suggested to check the residuals with different
    alignment constants after the calibrations are complete. Right now
    we are using eg1-dvcs alignment constants determined by Bo Zhao. 


[edit 
<http://clasweb.jlab.org/rungroups/lowq/wiki/index.php?title=Minutes_of_the_March_2_meeting&action=edit&section=2>]


    RTPC calibrations


    Yohann continued studies of *sdist* distributions, in particular the
    two peaks in *sdist* at some particular z ranges. 

        * He talked to Howard about this, who suggested that this might
          be due to the ALTRO chip clock. A shift in one clock pulse of
          100ns would correspond to a distance of 05.-1.0mm according to
          Raphael.

        * Yohann showed a plot of *sdist* versus ? integrated over all
          z. The plot showed acceptance-related ?-pattern of
          occupancies, but little or no ?-dependence of *sdist*.

        * Yohann also showed a plot of *sdist* vs z, which exhibits very
          abrupt and peculiar pattern. This pattern is most likely the
          reason for the two peaks since a finite z-bin size would
          produce 2 peaks at z~0. According to Howard this pattern is an
          artifact of /gem/ package.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/lowq/attachments/20100302/1195287d/attachment.html