Difference between revisions of "D2n Analysis To Do List"

Latest revision as of 12:01, 16 June 2010

General rules:

 If you find yourself searching for something up more than a couple of times
 then make a wiki entry with the information

 Try and keep the wiki organized sensibly
 - This might mean periodic consolidation of related information onto new pages
 - Keep the needs of others (and yourself) in mind when to add information
   - See if (related) information is already present in the wiki and add to it
   - Imagine looking at the entry 1 year from now and ask yourself:
     - will you know what it means?
       - explain/link to where the info came from, if relevant
       - if it only applies to a subset of runs, energies, etc, make that clear
   - Imagine trying to locate this information 1 year from now:
       - How will you be able to find it?
         - what terms will you search for?
         - have you saved the information to a logical place in the wiki?

Documentation

 Run Calendar
 - Starting point could be the RC calendar:
   http://hallaweb.jlab.org/experiment/E06-014/cal.html
   - Feb--Mar 2009 are the relevant pages
 - See: https://hallaweb.jlab.org/wiki/index.php/Analysis_resources_for_d2n#Calendar.2C_Run_Periods

 Target characteristics throughout the run
 - links to reference and 3He cell characteristics used during d2n
 - links to all raw target information taken during the run
 - table of 'processed' target information as a function of date/time
   - polarization direction (w/ error)
   - polarization value
     - w/ corrections applied

 Run database
 - target and beamline information should be integrated into run database
   - need to be careful that we don't get a mix of 'stale' data and 'new' data
   - perhaps have three independent databases:
     - beam characteristics     (needs to be consistent with db_run.dat)
     - target characteristics   (needs to be consistent with db_run.dat)
     - 'run' characteristics (incl. detector info)
 - make sure all relevant information is in the database
   - 'relevant' === stuff you find yourself scanning the log book for
 - make a 'how-to' in the wiki
   - add examples that you have found useful
   - add 'wish-list' items for things that should maybe be added

 Table of 'special runs' 
 - Data should be pulled from run database, but it will be useful to have
   convenient summary tables  for the following:
 - target 3He elastic runs
 - BB   optics/calibration runs
 - LHRS optics/calibration runs
 - BPM calibration runs
 - BCM calibration runs
 - BigBite Cerenkov LED/gain check runs

LHRS

 detector calibrations:
 - position determination from survey
 - pion rejector
   - energy calibration
 - Cerenkov
   - energy (# p.e.) calibration
 - timing, walk corrections?
   - timing most useful in early coincidence running to reduce backgrounds,
   - not sure if we have the resolution (or need) to extract particle velocity
     from ToF between sub-detectors
     - might allow for some small (negl?) improvement in background rejection

 Particle ID calibrations
 + Basic approach is to tune one of the two PID detectors, cut on a particular
   particle in that one then establish cuts on the second.  Iterate back and forth a
   couple times until the respective cuts are optimized and understood.
   - It will be simplest to do this process with a low p0 setting that has a
     good number of electrons and pions so you can see the effect of your
     cuts.
 - pion rejector
   - determine electron efficiency: (electrons detected)/(total electrons) vs. 
     - 'loose/tight' cuts
     - spectrometer momentum,
     - beam energy
   - determine pion rejection efficiency: 
       (pions rejected by cuts)/(pions that are mis-identified as electrons) vs. 
     - 'loose/tight' cuts
     - spectrometer momentum,
     - beam energy
 - Cerenkov
   - energy (# p.e.) calibration
   - electron efficiency
   - pion rejection ratio

 VDC/optics cross checks/calibrations:
 - "Bull's eye scan" BPM calibration
 - raster current <-> BPM calibration for raster correction
 - tracking efficiency checks
   - use 1-pass e,p elastic data and cut on proton arm so you know that an electron is 
 - track multiplicity measurements
 - sieve plate reconstruction
 - optics target (z-target) reconstruction

 Trigger efficiency
 - T3 vs. T4 trigger analysis

 LHRS Acceptance
 See 'Effective Use of Hall A HRS Acceptance With R-Functions'

BigBite

 detector calibrations:
 - position determination from survey
 - shower/pre-shower
   - energy calibration
 - Cerenkov
   - energy (# p.e.) calibration
     - start with second half of run period where Cerenkov was stable -- we'll
       work on the first half later
 - MWDC
   - plane-to-plane tracking efficiencies vs. luminosity
     - ie. vs. target, beam current
   - MWDC chamber position corrections via residual analysis
   - drift-time correction calibration
 - timing, walk corrections?
   - timing most useful in early coincidence running to reduce backgrounds,
   - not sure if we have the resolution (or need) to extract particle velocity
     from ToF between sub-detectors in single-arm
     - might allow for some small (negl?) improvement in background rejection

 MWDC/optics cross checks/calibrations:
 - "Bull's eye scan" BPM calibration
 - raster current <-> BPM calibration for raster correction
 - tracking efficiency checks
   - use 1-pass e,p elastic data and cut on proton arm so you know that an electron is 
 - track multiplicity measurement
   - we really want only single-track events, BB is noisy and will have a significant
     fraction of multi-track events:
     - what is 1:2:3:4:... track multiplicity for a given luminosity (and/or target?)
     - how does this improve when you apply geometric cuts on which blocks
       fired in the shower+pre-shower?
       - ie. project each track back to the shower plane and see if there is a hit 
         in the associated block(s)
 - sieve plate reconstruction
 - optics target (z-target) reconstruction

 Cerenkov cross checks
 - use software cuts on the Shower-only BigBite trigger and cross check the performance
   with the Shower+Cer geometric overlap trigger

 Particle ID calibrations
 + Basic approach is to tune one of the two PID detectors, cut on a particular
   particle in that one then establish cuts on the second.  Iterate back and forth a
   couple times until the respective cuts are optimized and understood.
   - You can start with a pretty wide momentum cut here, the low-p0 part will have
     a pretty high pi:electron ratio

 - pion rejector
   - determine electron efficiency: (electrons detected)/(total electrons) vs. 
     - 'loose/tight' cuts
     - spectrometer momentum,
     - beam energy
   - determine pion rejection efficiency: 
       (pions rejected by cuts)/(pions that are mis-identified as electrons) vs. 
     - 'loose/tight' cuts
     - spectrometer momentum,
     - beam energy
 - Cerenkov
   - energy (# p.e.) calibration
   - electron efficiency
   - pion rejection ratio

 BigBite Acceptance
 - "shouldn't" be a problem since it drops out in asyms calculation
 - would be interesting to place some tight cuts and see what kind of
   cross sections can be extracted...

Target analysis

 polarization calibration constant 
 - cross checks between EPR, water NMR signal
 - polarizing gradient between target and pumping chamber
 N2 dilution analysis
 3He pressure curve analysis
 1 pass elastic 3He polarization cross checks
 - vs. 'SAMC' monte carlo simulation
 Target density and temperature analysis
 Target Holding field direction (compass measurements)

Beam calibrations

 polarization
 - table/calendar with Moller results, half-wave plate changes
 - Compton 
 charge asym analysis
 BCM calibration
 raster calibration/correction
 Check for possible asym in gate timing: see Halog 280105

Compton analysis

 - Cross calibration between the Moller and Compton photon asym
 - linearity checks
 - IHWP consistency checks
 - correction for misbehaving power bit
 - cross calibration between original Saclay DAQ and new CMU DAQ
 - other systematics (beam position, beam charge asymmetry, ...?)
 - Compton asymmetry history for each beam configuration
 - error analysis
 - . . .

 Direct extraction from photon aysm + MC?

And all the rest... (to be filled out later)

• check for any bleed through effects from Hall C laser
• Livetime corrections
• Radiative corrections
• False asymmetry analysis

 - look at Carbon, N2, 3He (ref cell) data

• Positron backgrounds

 - simulation/model
 - 5- and 4-pass positron runs

• 3He to neutron

 d2n, g1, g2 extractions, etc.
 . . .

Milestones (Please add/update)

 detector calibrations (LHRS + BigBite)
 - show reconstructed sieve plate
 - show reconstructed z-target for optics

 particle ID calibrations
 - plots of pion-rejection and electron efficiency vs. PID cuts

 tabulate "run-sets" for focused analysis
 - blocks of runs that you will need 'special' treatment
   - ie. different beam energy
   - some change to a sub-detector
     - ie. change in hardware, etc.

 LHRS specific
 - 12C(e,e') cross section extraction and compare to existing data
 - 3He(e,e') abs. cross section extraction and compare to existing data

 BigBite specific
 - extract prelim. asym in medium momentum bin (ie. not too low cross section,
   but not too large pion contamination)
   - show it changes sign with target flip
   - show it changes sign with IHWP flip

 Compton calibrations
 - power bit corrections
 - IHWP sanity checks
 - Compton/Moller cross-calibration over course of experiment
 - Saclay/CMU DAQ cross-checks for each beam configuration
 - Compton history plots for each beam configuration