Apr 23-24/26 PionLT/KaonLT Analysis Meeting Notes ------------------------------------------------- (Notes by GH) Today: KaonLT will be discussed first Thursday: Present ----------------- Regina - Garth Huber, Nathan Heinrich, Nermin Sadoun, Nacer Hamdi, Alicia Postuma CUA - Tanja Horn, Chi Kin Tam, Sameer Jain Nacer ----- KaonLT Q2=0.5 LT-sep - solved yield problems reported last week (drop in high epsilon event counts) - some files used in new replay were incorrect, events were skipped - still waiting for final replay jobs to finish Chi Kin ------- KaonLT High Q2 LT-sep Q2=3.0 W=3.14 - ran more iterations after applying t-shifts, still using old optics ME - kinematic distributions looking better now. a selection of data vs MC comparison plots were shown, comparison now looks good for all 5 SHMS settings (both epsilons), while before only 3 looked good - working on implementing Gabi's method to optimize kinematic offsets from physics data - waiting to rerun Sigma0 SIMC events - *NB* some questions were raised on what beam energy offset is being used in the simulations and replay, CKT should consult with Richard on this Alicia ------ u-channel analysis - working on setting up code for LT-sep iterations - started from Junaid's code, need to add the u-channel-specific background subtractions - documenting things as work progresses - looking at Bill's old code to see what checks were done, etc., but applying to new analysis framework Sameer ------ PionLT Q2=1.6 W=3.08 background subtraction - MMp plot vs NGCer.npesum shows evidence of electrons caused by pion knock-on - *NB* Nathan: suggest to look at MMpi vs PCAL region near 1, that should remove the knock-on electron events - using NGC cut for Dummy target - formed CoinTime random subtraction windows - had some questions for group on how to calculate Qeff - *NB* should use HMS Calorimeter and Cer efficiencies from Heep data analysis, don't recalculate these from physics data, as there will be too much pi- leakthrough to get a reliable result - other factors to include: TrackEff, LiveTime, pion abs corr, LH2 boiling corr from Nathan - Nathan: for detectors other than HMS Cal & Cer, the efficiency calculation cuts need to be the same as for the physics data - The tracking efficiencies depend on the tracking algorithm and the particle type, but they should not be particularly sensitive to exact details of used PID cuts (on other detectors) - *NB* CoinTime Blocking needs to be tabulated in report files and applied run-by-run Friday: Present --------------- Regina - Garth Huber, Alicia Postuma, Nathan Heinrich, Nacer Hamdi, Vijay Kumar FIU - Pete Markowitz JLab - Dave Gaskell JMU - Gabriel Niculescu CUA - Chi Kin Tam Virginia - Richard Trotta Glasgow - Rachel Montgomery Ohio - Julie Roche Richard ------- KaonLT High Q2 LT-sep - decided against doing a new Pass-4b replay - eneded up re-running the efficiency script on Pass-4a replay data, checked some CoinBlock values and they looked good - will do a Pass-4c reply with different HMS optics ME - Pass-4a has phi-offset=0 and original 2018 ME - new replay will have new 2018 ME and phi-offset from https://github.com/JeffersonLab/nps_replay/blob/dae2ba5d2db619dc63cd46b9ce5c1f67414b7a96/PARAM/HMS/GEN/hmsflags_opt2018.param - plans to start new replay next week - *NB* questions raised on whether Richard's empirical fit method to remove background underneath Lambda peak is still justified, it may be removing too much Lambda radiative tail - pion leakthrough w/ Chi Kin's new HGC hole cut is reduced - Q2=3.0 W=3.14 hi-eps data MM plot is particularly clean, no evidence of background to left of Lambda peak - Nathan: raises concerns on whether there is any remaining pion background to subtract at all - plots shown for other settings have substantial #events to left of Lambda peak, but empirical fit seems poorly justified - *NB* Gabriel: should compare the MMk spectrum to SIMC Lambda lineshape before applying empirical fit, to be sure not too much radiative tail is removed - *NB* Garth: reminds everyone that SHMS collimator rescattering routine in SIMC only is active for pions right now, the code should be modified for p,K, as without it SIMC is likely under-estimating the size and shape of the Lambda tail - Alicia plans to work on this for proton u-channel analysis, can include also for K+, will discuss this with Garth offline - *NB* Nacer: an empirical fit to the full MM region may be unnecessarily complicated, suggests to just fit the region around Lambda before looking at Sigma0 region Vijay ----- PionLT Low Q2 LT-sep - mostly busy with paper - made a table summarizing systematic uncertainty studies for Q2=0.375 - long row of columns of %changes in separated xsec vs t-bin - *NB* Garth will take a look and discuss with Vijay offline on how to use this info to compute systematics Nathan ------ PionLT LT-sep - Q2=5.0 W=2.95 mid-eps left-SHMS analysis - using Nacer's PID code, modified for pion sutdy - plot of RFtime vs MM clearly indicates it cannot be used to separate pi,K for this setting, the time offset between them is too small to be useful - shown CoinTime plot is not properly centered at 0, cuts are off - using ROC1, need to switch to ROC2 - no Aerogel cut to remove protons is applied yet - Q2=6.0 W=3.19 mid-eps - replay just finished, showed fresh plots - CoinTime has a double peak structure, it seems that somewhere in the ~100 runs of this setting the CoinTime offset changed - still lots to look at here Next Meetings ------------- - Thurs: Apr 30 @ 16:00 Eastern/14:00 Regina - PionLT will go first - Fri: May 1 @ 11:00 Eastern/9:00 Regina - we will continue where we left off Going forward: - the Friday morning meeting conflicts with the new Hall C RC meeting time, so we need to move at least the Friday meeting - please fill out the poll at: https://www.when2meet.com/?36401043-97RnM