[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: calorimeter document




Hi Elke,

According to my notes from our phone call the other night, I have made  
the following changes:

- removed my internal flag about "need to check consistency of SiPM  
parameters" and added the sentence:

Based on the design specifications for the SiPM devices, the following  
method was used to estimate the electronics threshold.

- explaining the background shapes in pi^0 and eta just got to be too  
vague and cumbersome to be useful so I didn't do it

- tried to make clear that we are rejecting only pythia generated  
channels that exclusively match our signal

To ensure that the sample is originates from pure background hadronic  
photoproduction processes, we examine the identities of the generated  
particles as provided by \pyth{} and discard events that match our  
{\em exclusive} signal final state topology.

- sprinkled in a few more sentence around the dense text about our  
"poor man's" kinematic fit:

The most effective method of suppressing remaining hadronic  
backgrounds is to employ a global kinematic fit that enforces four- 
momentum conservation and evaluate the goodness of fit for the signal  
candidate; however, such techniques have not yet been fully  
developed.  In lieu of a global kinematic fit, we simply place  
requirements on the total reconstructed energy and transverse momentum  
of the signal photons.  Figure~\ref{fig:e0pyt_etot} shows the total  
energy of the $\eta\pi^0$ candidate photons for events with only one  
candidate from \pyth{} background (top left) and from signal $\eta 
\pi^0$ (top right).
The red points in the top right plot show the background from \pyth{}  
scaled to the signal based on the ratio of cross sections predicted by  
\pyth{}.  In order to mimic the energy conservation constraint that a  
full kinematic fit would enforce, we select events with $E(4\gamma) >  
8.0$~GeV.  In addition, the total transverse momentum $p_{\mathrm{T}}$  
must sum to zero to conserve momentum.  The lower pair of plots in  
Figure~\ref{fig:e0pyt_etot} show, in the same style, $p_{\mathrm{T}}$  
of the signal $\eta\pi^0$ candidate and recoil proton after the  
$E(4\gamma)$ requirement.  To enforce conservation of momentum, events  
with $p_{\mathrm{T}}<100$~MeV/$c$ are selected.

and added to the end of the paragraph:

For the analyses presented here, we make no additional selection  
requirements although it is recognized that there exist additional  
techniques for further suppressing background, {\it e.g.,} selecting $ 
\pi^0$ or $\eta$ candidates whose decay photon energies are not  
heavily asymmetric.

Unfortunately I just do not have time to complete the following things:

- alterations to conversion figure to indicate detector regions where  
conversions happen (it is noted in the caption)
- addition of Beni's rainbow scatter plot showing conversions
- addition of summary table for simulated and generated properties

I will certainly include the first two in my talk.  I don't think the  
last is a disaster because the only place we could make a direct  
comparison between actual beam test and simulation would be in BCAL  
energy resolution.  The remainder of the items are either implemented  
in simulation and not completely independently verified or measured  
(like BCAL timing resolution) but not implemented in simulation.  Of  
course, I realize there is some value in collecting the numbers all in  
one place.

Cheers,

Matt