[Primexd] [New Logentry] Initial CCAL (NPS prototype) calibration and energy resolution

[somov at jlab.org]

Logentry Text:
--
I )  In order to equalize gains on CCAL modules 
we performed a HV scan for each module (runs 
51824 - 52004). Each CCAL module was positioned
to the photon beam. Energy of beam photons was
determined using tagging detectors. HVs were 
computed according  to the required energy range:

- energy deposition of a 10 GeV photon 
is required to be around fadc count 3100  
(note, the maximum fadc range is 4096 counts)

- we set voltages by requiring the average fadc 
amplitude for 4.2 GeV photons to be 1267 fadc 
counts. 

--------------

After setting voltages, we checked the calibration 
(runs 2216 -22364)

Fig 1. FADC amplitudes of CCAL modules corresponding 
to 4.2 GeV photons. Sigma of the distribution is
about 1.3 % 

Fig. 2 FADC amplitudes of CCAL modules corresponding 
to  ~10. GeV photons

Fig. 3.The amplitude ratio (A (10 GeV) / A (4.2 GeV)
The relative width of the distribution is about 0.4 %
(relatively stable performance of all modules).


The HV setting procedure worked well for most 
modules except of 10:

We didn't calibrate two modules (5,4) and (-2,-6).
There are 7 suspicious modules (-4,6), (-2,6), (-6,5),
(-4,4), (4,-2), (-6,-4), (5,-5).

For these modules, we adjusted voltages "by hand".
We'll check calibration after we finish with
Compton production runs.


-----------

II ) We used the initial gain calibration to estimate
the energy resolution of CCAL


Fig. 4 Energy distribution for 4.2 GeV photons

- energy deposited in the single CCAL module (curve on the left)

- total energy deposited in 5x5 cells (curve on the right). The 
  relative energy resolution (width of the distribution over the 
  mean value is about 2%)

We checked different regions in the detector. The typical 
energy resolution for 4.2 GeV photons is 2 - 2.3 %


Fig. 5 Energy distribution for 10 GeV photons

- the typical energy resolution is 1.4 - 1.6 %


The energy resolution can be improved after we refine gain 
calibration. Two things have to be done: 

- gain can be refined by fitting energy distributions in 5x5 
  modules to the shower profile  

- data have be taken in a raw fadc mode 
  (in Fig. 3 and Fig. 4 the fadc threshold for each module 
   was set to 27 MeV)



---

This is a plain text email for clients that cannot display HTML.  The full logentry can be found online at https://logbooks.jlab.org/entry/3639399
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/primexd/attachments/20181216/8db28176/attachment.html>