Kaon_LT Experiment

                 Feb 8/24 PionLT/KaonLT Analysis Meeting Notes

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

                                (Notes by GH)

                     Today: PionLT will be discussed first

Please remember to post your slides at:

https://redmine.jlab.org/projects/kltexp/wiki/Kaon_LT_Meetings

Present

-------

Regina - Garth Huber, Ali Usman, Muhammad Junaid, Nacer Hamdi

CSULA - Konrad Aniol

York - Stephen Kay

JLab - Dave Gaskell

CUA - Richard Trotta, Tanja Horn

Junaid

------

Errors on Tracking Efficiencies for Heep singles data

- found the error in the calculation

- data is replayed, will be looking at the results soon

Richard

-------

SIMC Weights

- last week's issue was due to weight not being applied to SIMC yields

- with this fixed, the 10^6 factor is no longer needed

- now getting Data/MC ratios of 0.2-0.5

SIMC Missing Mass

- last week reported a very long K+Lambda radiative tail, the opposite that was

  found for the pi+ analysis

- Dave looked at RT's input files, and found that the wrong K+ decay length was

  used

  - 780.4cm was used instead of 371.2cm (i.e. pi+ value instead of K+ value)

  - SIMC would have generated the correct K+ decay products, but with the wrong

    probablity

  - the fact that the user has to input the correct decay length (instead of it

    being selected automatically since SIMC already knows it's handling K+

    decay) is a historic artifact from Dave Koltenuk days

- a second issue was that MM was not recalculated for SIMC events by the

  recon_hcana_script

- with these corrected, the K+ radiative tail agrees better with the data, of

  course the data are still subject to change as well, until the PID cut study

  is completed

K+ PID Studies

- investigating pi+ and proton leakthrough into K+ sample

- selected three different MM regions:

  MM<1.05 (should be mostly leakthrough protons)

  MM>1.16 (will contain leakthrough pi+ as well as real K+)

  K+Lambda region is 1.06<MM<1.16

- for MM<1.05 region, look at Aerogel NPE distribution

  - saw some evidence of proton leakthrough

  - tightened cut

    before: events with >3NPE accepted

    now:  events with >4.5NPE accepted

  - still need to adjust HGC hole cut

    - HGC optics were adjusted between low and high epsilon runs, need larger

      hole cut for 10.6 GeV data

- MM>1.16 region, again look at Aerogel

  - will try 4.5NPE cut, maybe need 4NPE if cut is too restrictive

  - need to apply same Aerogel cut to all MM regions

Data/MC Yield Comparisons

- looking at Q2=2.115 yields with different MM cuts

- ratios are closer to 1 with 1.06<MM<1.16 cut applied

Separated Xsect Comparisons

- iteration scripts seem to work

- some discussion of what units used for cross-sections in SIMC, Richard still

  doing final checks

Next Steps

- finish PID studies

- finish cut analysis

Ali

---

Compare pi+n MM dists of Data and SIMC with Dave's delta fix

- Bad News: somehow the Data MM distributions before and after delta fix are

  identical, which is suspicious.  Will discuss w/Alicia after she's back

  - Plots for Q2=5 are shown, others behave the same way

- Good News: adjusted Drift Chamber resolution parameter in SIMC

  - with the adjustment, the pi+MM peak agreement between Data and SIMC is

    significantly better

  - only adjusted the SHMS resolution so far, since it's higher rate

  - HMS routine already had some resolution tuning (apparently from Doug

    Koltenuk), while SHMS lacks this code

- Dave: Abideh will look at HMS inclusive data and tweak HMS resolution there.

  He can provide his HMS parameters when he's done, so the effects of the two

  spectrometers can be separated that way

- Dave: the DC intrinsic resolution should not be strongly rate dependent.

  However, at higher rates, the Gaussian tail fraction does increase.

  - Koltenuk's HMS resolution tuning is along this idea, with a double Gaussian

    distribution giving bigger tails

- Dave: when comparing SIMC to Data MM distribution, try to confine your study

  to the region just around the pi+ peak, to avoid attributing DC resolution

  issues to what might be problems in the treatment of the radiative tail

  - suggest to look at +/-2 sigma region around pi+ peak, so MM<0.98

  - really hoping the SHMS DC resolution parameter does not turn out to be

  kinematics dependent.  If so, this would indicate that other parts of the

  simulation (such as multiple scattering) are incomplete

Nacer

-----

KaonLT low Q2 DC calibs 95% done

- submitted last jobs this morning

Next Steps

- will need to replay all data with new DC calibs, and offsets

- then can start looking at K+ PID

  - Note: Aerogel index (n) will be higher than in Richard's data

Stephen

-------

Notes that starting a new WorkFlow on SWIF seems to give a better success rate

on iFarm jobs

- Nacer notices the same effect

Next Meeting

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

- Thur Feb 15 @ 15:00 Eastern/14:00 Regina

  - KaonLT will go first