[d2n-analysis-talk] Dimensions of BigBite Preshower and Shower

[Brad Sawatzky]

On Wed, 06 Apr 2011, Diana Parno wrote:

[ . . . ]
> Finally, I looked at our DB files in /usr/local/d2n_analysis/d2n/DB/ 
> db_BB.ts.ps.dat and /usr/local/d2n_analysis/d2n/DB/db_BB.ts.sh.dat. It  
> seems we're using a completely different number there: 35 cm for the  
> long dimension of each block. Here are a few relevant lines from the  
> preshower file:
> 
> Half of X, half of Y, full Z sizes (in Meters) of Preshower
>      1.1475  0.35   0.085
> dx and dy block spacing
>     -0.085   0.35                                 - Meters
> 
> And from the shower file:
> 
> Half of X, half of Y, full Z sizes of Shower
>     1.1475   0.2975   0.35                        - Meters
> dx and dy block spacing
>     -0.085  0.085
> 
> Recall that X is vertical, so a vertical half-extent of 1.1475 m is  
> consistent with 27 rows of 8.5-cm blocks.
> 
> So we have here three numbers: the long edge of the blocks could be 34  
> cm (Transversity), 35 cm (our DB file), or 37 cm (GEn documentation).  
> I wouldn't have expected the calorimeter to be disassembled and  
> reassembled with slightly different blocks so many times in the space  
> of a few years; I suspect there is one right answer, but at this point  
> I'm not sure how to determine conclusively which one it is. So, I have  
> two questions:
> 
> 1. What are the actual dimensions of the BigBite shower and preshower
> blocks?

Bogdan's the authority, so 8.5 x 8.5 x 34cm it is.

> 2. If they are different from 8.5 x 8.5 x 35 cm^3, then we have the  
> wrong dimensions in our DB file. What effect does that have on our  
> analysis?

It should certainly be corrected, but I wouldn't think it'll make any
difference.  The shower is the most "dimension-sensitive", since it
provides coordinates that are cross-checked against track.  The x,y
dimensions are correct for the shower.  The depth is off by 1 cm, but
I assume the blocks are positioned inside the full stack relative to
their upstream face.  If that's the case, then they're just a little
short -- might matter a little for a Monte Carlo, but it should affect
the data analysis.

The preshower assembly is 2cm wider that it should be (it'll stretch
outside the shower solid angle), but I don't think that dimension is
actually used in a way that would impact the analysis.  It would be a
small problem if the DB block dimension was too small, since that might
impact the effective solid angle built in to the shower hit-position
code.

So, I guess I'm not too worried.  It would be good to replay a run-set
with and without the updated numbers and then generate a few plots
for comparison.  The '# particles that pass each cut' would be a good
set, as would the shower-based position histograms.  Since the same
files are used in each replay, it is deterministic and statistics won't
be a big issue (a short replay should be sufficient).

Matt, would you be willing to do this check?  (Or Diana, if you have the
scripts handy?)

-- Brad

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