Meeting to discuss Kaon-LT Efficiencies for Normalized Yield
                               22-Jun-02

                        (Notes by GH and SJDK)

Attendees: Stephen Kay, Garth Huber, Ali Usman, Richard Trotta, Vijay Kumar

Ali Usman
---------
- shows Heep COIN data/MC comparo using Richar's script

- Richard tried to automate the charge normlisation
   - Didn't seem to work particularly well
   - Data ends up far smaller than SIMC
   - Manual vs automated calculation
      - Dummy charge is almost exactly correct, since it's only one run
      - Data number is way off
         - 40% of manual calculation
      - RT - Probably got the reason, hadd doesn't do scalers correctly
      - AU's manual calculation was just manually grabbing numbers from report
        file for 4 files, so didn't have the hadd problem
   - GH - For consistency, *should get charge from the report file*
      - Efficiencies script can just grab this anyway

- New cuts file with Singles and Coin track efficiencies
   - All events now processed, so will take longer to process
   - SK - Test a very large file, how long does it take?
      - I'll try and find a very large file
   - the output ROOT file should only contain "loosely good" events, to save
     space and processing time.  GH calls this "skimmed data"
      - Skim cuts:
         - good COIN events w/tracks in both spectr
         - good delta cuts on both spectr
	 - No PID cut on SHMS, loose HMS-CAL cut for electrons
      - GH: Need to be careful when including detector cuts, though 
         - e.g. When was the HMS cherenkov mirror fixed?
         - Had a large crack across it, completely broken
         - Was replaced June 2018 - https://logbooks.jlab.org/entry/3576642
	 - *RT will make a plot of HMS eff vs (x,y)* [similar to what is
            done for HGC] from the skimmed data, to make sure all is okay

- a 2nd explanation for the narrower Pmiss, Emiss peaks by SIMC compared to
  data that Ali sees could be that the assumed beam energy spread in the SIMC
  input file is too small 
   - Arc energy measurement doesn't include energy spread -> Check with DG on
     *where we can get energy spread info*
      - Beam energy width broadens with pass # and beam energy
         - Wider as a percentage of a larger number, so it's a lot,
           particularly for 5-pass 10+ GeV beam
         - Someone will need to chase down someone at JLab regarding this
         - An example log entry -  https://logbooks.jlab.org/entry/3909335

Richard Trotta
--------------
- Was seeing randoms in Heep coin time plot
   - *need to place cuts around prompt peak and do random subtraction*
   - also need to investigate MM, to see if there is an inelastic contribution
     at higher energy that needs to be cut out

- After scaling MC data, shapes actually look OK on HeeP data
   - Just need to sort out the charge normalisation
   - Shapes of missing momentum are very odd
      - Discussed in meeting yesterday (KaonLT meeting)
      - Eqns between SIMC/hcana are different for missing momentum etc
         - RT/AU probably have to fix this, but it's better to try simpler
           solutions first, such as the beam energy spread
      - Vijay sees similar effect but much smaller due to lower energy

- Efficiencies script
   - grabs kin from standard.kinematics
   - Cer Effs, from Vijay's script, but not yet incorporated
   - pi-Abs corr and Target Info, to do
   - Time added in csv script
   - Keys in dict
      - Just need new variables from Ali
   - GH - Would be nice to have a quick demo w/o the variables you're waiting
     on from Ali

- Plots for Vijay's Hall A/C talk
  - RT will make some plots of Effs vs Run #
  - AU will finish template files
  - Aim to reprocess KaonLT replay (all data over the weekend)

- AU raises issues with the *Calorimeter Effs*
  - reminder that everyone needs to do a study for their assigned run periods
  - one person can do a study for their run period, and others then just
    optimize the cuts for theirs, and not start from scratch

Next meeting:
- Mon Jun 13 @ 15:00 Eastern