Re: [Beamtest Analysis] - Attn:Dave and Simon - odd tphoton spectra(fwd from David)

[Elton Smith <elton@jlab.org>]

Hall D barrel calorimeter:


---------- Forwarded message ----------
Date: Mon, 27 Nov 2006 10:24:41 -0500
From: David Lawrence <davidl@jlab.org>
To: Simon Taylor <staylor@jlab.org>
Cc: Blake Leverington <leverinb@uregina.ca>, halld-bcal@jlab.org
Subject: Re: [Beamtest Analysis] - Attn:Dave and Simon  - odd tphoton
    spectra


Hi,

    Just to add a little to Simon's comment. The production trigger(bit 4 for run 2334) had the
MOR in it which I believe was delayed so that it determined the trigger time. This means that the
jitter between T-counter channels would appear in the trigger time.

    I agree that we'd like to have nicely lined up timing spectra, however, I want to make sure we
don't get caught chasing a red herring here. The only thing we really need good timing from the
tagger for is to match it up with the event seen in the downstream detectors (veto and BCAL). My
understanding is that generally, you apply a cut that Nphotons==1 so there is no ambiguity as to
what photon caused the event. Thus, your cuts on the tagger timing can be very loose. In this case
+/-10ns.

    The most vital information about the BCAL comes from the BCAL itself. If I look at the time
difference between N8 and S8 for the recent 2334 DST, I get:

root [53] bcal->Fit("gaus","tn8-ts8","abs(tn8-ts8)<10 && Nphotons==1")
 FCN=13346 FROM MIGRAD    STATUS=CONVERGED      94 CALLS          95 TOTAL
                     EDM=8.04063e-09    STRATEGY= 1  ERROR MATRIX UNCERTAINTY   1.9 per cent
  EXT PARAMETER                                   STEP         FIRST
  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE
   1  Constant     5.29003e+04   1.55913e+02  -6.24633e-01   1.42340e-06
   2  Mean        -1.11402e-01   8.19901e-04   1.57801e-06  -4.41226e-02
   3  Sigma        3.86152e-01   7.91046e-04  -9.46867e-07   4.28185e-01

Which says the timing difference has a sigma of 386ps (which includes resolution due to spot
size). If you want to look at absolute timing resolution of the BCAL relative to an external
detector, you should place cuts on Ephoton such that you're looking at a single T-counter.

Regards,
-David


Simon Taylor wrote:

 Hi, Blake.

I believe the explanation is this:
The last timing calibration for the tagger counters was done relative to
signals in the veto (using one of the special veto-triggered runs).  The
times you plot are relative to the (presumably) production trigger.  The
T-counters are not guaranteed to be exactly in time in hardware at the
trigger level.

Simon

On Fri, 24 Nov 2006, Blake Leverington wrote:



 Hi Dave and Simon,

Hope you guys had a good weekend stuffing yourselves!

I've been doing some more analysis looking at the tdc and so on to try
and clean up the adc spectra for the BCal. I noticed when I plot adcsumN
vs. tphoton and Ephoton vs. tphoton I get the plots at the bottom of
http://www.jlab.org/Hall-D/software/wiki/index.php/BCAL_Beam_Test_Plots%2C_November_20%2C_2006

The scale is blown up but I think something strange is going on like the
offsets aren't correct for the tagger. Or is this normal and something I
have to live with?

Thanks,
-Blake




 --

------------------------------------------------------------------------
  David Lawrence Ph.D.
  Staff Scientist                 Office: (757)269-5567   [[[  [   [ [
  Jefferson Lab                   Pager:  (757)584-5567   [  [ [ [ [ [
  http://www.jlab.org/~davidl     davidl@jlab.org         [[[  [[ [[ [[[
------------------------------------------------------------------------