Difference between revisions of "Online analysis: results and TODO"

From Hall A Wiki
Jump to: navigation, search
(Matching data taking with run plans)
(Cerenkov Efficiency Study)
Line 32: Line 32:
 
Prior to afternoon of 1 Nov 2010, Cerenkov threshold at 200 mV.
 
Prior to afternoon of 1 Nov 2010, Cerenkov threshold at 200 mV.
 
Inefficiency = 7%
 
Inefficiency = 7%
 +
After lowering the threshold to 50 mV, Cerenkov inefficiency < 1.5%
 +
(Updated 12 Nov 2010)
  
[[Image:CerenkovTrigger-v1.pdf]]
+
[[Image:CerenkovTrigger-v2.pdf]]
  
 
== Deadtime==
 
== Deadtime==

Revision as of 14:20, 12 November 2010

Back to DVCS

This page is in progress, please new items to it. When you obtain a result, log in into the Halog and link it to this page. if you are willing to take charge of a given analysis, please indicate it on this page as well as the date at which you are starting to work on it.

Matching data taking with run plans

  • Update the how to
  • Maintain db_run.dat file: Make sure that every LHRS momentum and angle changes are logged (Florian Itard is keeping up with that)
  • Database of run indicating line by line the run number, the target in position, the average beam current, LHRs pnom, LHRS angle, the purpose of the run

[Statistic summary]

Luminosity, efficiency

Boiling test

Initial calibration on 10-17 owl shift: [shift summary][initial plan]
Second calibration on 10-21 swing shift: [run list]

Trigger efficiency

Online analysis by B. Michaels [here]. 

Run7711 T3 and T4 Triggers

S0 hit pattern vs x and y (at vdc) shows efficiency in excess of 99% except at extreme ends of focal plane halog entry 336817 http://www.jlab.org/~adaq/halog/html/1010_archive/101026122751.html

Cut script saved at /adaqfs/home/adaq/dvcs/onlana/GoodHRS.h TCut cGoodVDC = "L.tr.n==1&&abs(L.tr.x)<1."; TCut cGoodAngles = "abs(L.tr.tg_th)<0.1&&abs(L.tr.tg_ph)<0.05"; TCut cTarget = "abs(L.tr.tg_y)<0.1"; TCut cCleanTrk = cGoodVDC&&cGoodAngles&&cTarget;

Cerenkov Efficiency Study

Prior to afternoon of 1 Nov 2010, Cerenkov threshold at 200 mV. Inefficiency = 7% After lowering the threshold to 50 mV, Cerenkov inefficiency < 1.5% (Updated 12 Nov 2010)

File:CerenkovTrigger-v2.pdf

Deadtime

LHRS commissioning

Calibrate the ADCs:

Still to be done S1, Pion rejector, (Shloka is working on that)

[Cerenkov], [S2 (done 10-09)]

Calibrate detector time spectra:

Still to be done : understand timing spectra for S1,S2 and the Cerenkov

Evaluate PID efficiency:

LHRS optics run (no sieve/sieve)

Ye Zhihong is interested in this analysis and started it.

LHRS acceptance check

[delta scan on 10-22] need to be analyzed

DVCS Calorimeter

  • two photon invariant mass: [10-24]

Elastic calibration

Beam

BPM calibration

BCM calibration

  • second calibration on 11-05 (after temperature is fixed): [here]
  • the initial calibration was done on 10-14 [halog]. Calibration results are [here]

Beam energy measurement

  • the initial measurement was done on 10-15: [halog]

Helicity

  • Helicity not defined gate: [here]
  • Helicity crisis: [the problem], [solution?]
  • Large charge asymmetry : helicity sign check: [this halog].
  • ring map file
    • November 1: new configuration [here]
    • September 20: ring map file [here]
  • runs with TSettle as trigger:
    • 7387: after the new helicity board is installed: [here]
    • test at various helicity flipping rate: [here]

Polarization

Analysis outputs

From Carlos:

You can find under the directory /adaql4/work1/dvcs/ntuples/ in the adaq computers ntuples corresponding to a first "fast" analysis pass.

Each file contains 2 ntuples (ntu1 and ntu2) corresponding to 1 and 2 clusters events. The name of the variables should be self-explanatory, but feel free to ask me.

These files correspond to a first analysis pass with a software calorimeter threshold of 1 GeV, fast waveform analysis (1-pulse fit only) and (elastic) calibration coefficients from Oct 26. Timing cut is -3,+3 ns wrt HRS. Additional cuts include 1-track only in the HRS and reasonable vertex reconstruction (within the target length).

All runs (even junk or LeftHRS-only, see PS below) are analyzed automatically as they are produced (although there is a backlog of a couple of days).

Please report any problems you might see, or any interesting variables missing.