[d2n-analysis-talk] T2,T6 time walk

[Brad Sawatzky]

On Tue, 23 Mar 2010, MATTHEW R POSIK wrote:

> Here is the link to the plots that I refer to for the t2 timing:
> 
>     http://jlab.org/~posik/d2n/BB/trigger/t2_walk/t6_timing/
> 
>  I took a look at the timing on the TDCs. Your explanation makes a lot
>  of sense. I plotted a corrected TDC 03 to carry the T2 timing and
>  also show the TDC with no T2 timing correction (corrected_TDC03.png
>  and uncorrected_TDC03.png). As you can see the peak is much narrower
>  and has more events under it. But there seems like there may still be
>  a little bit of a shoulder to the right of the dominate peak. This
>  may be related to the self-timing peak  that I chose. There seems to
>  be more than one peak in the DBB.t2 trigger (t2.png), I selected 560
>  as my self-timing peak, but I am not sure that this selection is
>  correct.

I will admit that I don't really understand the DBB.t2 histogram.  Self
timing spikes are typically 1--2 bins wide -- ie. more like the spike at
bin 200 than the ones out >500...

The Cerenkov TDC peak does look better though, so the spikes at >550
are clearly measuring the T6-T2 time difference at some level -- maybe
I'm just getting spooked by the log-scale (it highlights the tails on a
gaussian and makes peaks look wider than they should)...

Hmm.  I'd like to understand where all that structure is coming from
though.  What run are you looking at?

> As for the tracking variable, I thought I had fixed this too. I
> realized this problem when looking at some physics replays for the H2
> elastic events.  To correct it I ended up removing some cuts from the
> .cdef file. But the cdef file that I am using now is empty. And when
> looking at the H2 elastic events that I replayed yesterday on my
> current d2 set up I get more one-track than 0-track events. Where as
> when I replay a 4-pass 3He and a 5-pass N2 run I get more 0-track
> events. These can be seen here:
[ . . . ]

Diana, could you replay this run (Matt will tell which one) and see if
you get the same numbers?

> Any suggestions on what may be causing this? Could the mwdc be too
> tailored to the transversity settings?

I don't think it should be too bad -- the optics didn't change that much
(as indicated by Xin and Diana's plots from some time back).  There's
some distortion at the edges of the acceptance that we can fine tune
out, but it wasn't horrible...

Are the plots you showed for T2, T6, or all triggers?

IIRC, 50% no-track events for BigBite T6 triggers might be about right.
There are a lot of gammas that fire the shower, but don't leave tracks.

T2 triggers involve the Cherenkov (and so require a charged particle).
There should be a much lower fraction of no-track events for that
trigger type.

The time-zeros for the chambers would have changed between transversity
and d2n though (since the trigger changed).  That is one of the
calibration constants in the DB files.  If those are only 'close' then
subsequent cuts in the tracking code might throw out the bulk of the
events.  The remainder (ie. tail of the peak) would still give a good
track and reasonable reconstructed momentum though.  It would be good if
you and Diana double checked those coefficients.

-- Brad

-- 
Brad Sawatzky, PhD <brads at jlab.org>  -<>-  Jefferson Lab / Hall C / C111
        Ph: 757-269-5947        -<>-        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