[d2n-analysis-talk] Fwd: LHRS Scintillator Calibration Study

[Brad Sawatzky]

On Fri, 05 Mar 2010, David Flay wrote:

> On Fri, Mar 5, 2010 at 11:04 AM, Brad Sawatzky <brads at jlab.org> wrote:
> 
> > Vince pointed out the answer.  Your cuts are working correctly --
> > the second band are cosmics going through the stack 'backwards'.
> > Note that for good electrons (bulk of the data) t_2m-t_s1=6ns, but
> > for the second band: t_2m-t_s1=-6ns.  At high p0 we have so few
> > electrons that cosmics actually compete.  Interesting.
> 
> Hmm, that is interesting.. So, I would need a cut on beta > 0 to make
> this plot look normal again?  I'll try it out..

Think about your logic here.  The _only_ variable that goes into beta is
the deltaT between S1 and S2m.   It is worse than useless to use cuts on
beta to 'clean up' the S1, S2m plots.   It's precisely equivalent to
cutting on S1 to clip out the extra band and then relaxing because the
band is gone.  That logic is completely nuts :-)

Do not look at beta at all until you understand the timing.

> regarding the S1 timewalk -- I plot left TDC vs. 1/sqrt(left ADC) and
> I get a straight line (using my corrected values -- to check them.)
> (based on Chiranjib's approach in the transversity ELOG)
> 
> This is good. -- if I <do not> have a cut beta > 0, I see a similar double
> band structure (separated by about the same time as seen in the avg. time
> plot) in this plot.

Of course.  The reason for this should be clear.

-- Brad

-- 
Brad Sawatzky, PhD <brads at jlab.org>  -<>-  Jefferson Lab / Hall C / C111
  Ph: 757-269-5947 -<>- Pager: 757-584-5947 -<>- Fax: 757-269-5235
The most exciting phrase to hear in science, the one that heralds new
  discoveries, is not "Eureka!" but "That's funny..."   -- Isaac Asimov