Feb 16/23 PionLT/KaonLT Analysis Meeting Notes ---------------------------------------------- (Notes by GH and SJDK) Today: KaonLT will be discussed first Please remember to post your slides at: https://redmine.jlab.org/projects/kltexp/wiki/Kaon_LT_Meetings Present ------- Regina - Stephen Kay, Garth Huber, Ali Usman, Alicia Postuma, Nathan Heinrich, Vijay Kumar, Love Preet, Muhammad Junaid JLab - Dave Gaskell Ohio - Jacob Murphy, Julie Roche CUA - Richard Trotta, Tanja Horn CSULA - Konrad Aniol Richard Updates --------------- Changes to PM/EM calcs in SIMC - Debugged ROOT issue, was overwriting old histos - DG: Script should write new tree and retain old tree, so you can compare them directly - Results have some unexpected featuers, will discuss with DG in more detail tomorrow morning LT-sep code - rewriting the part of code that tabulates the normalized yields, rathern than debugging Bill's version - GH: yes, this makes sense. The yield calculation is experiment specific, and is best to be written from scratch - it doesn't matter which variable is integrated to determine the yield, but as Vijay found earier, it's good to try integrating more than one variable as an initial test, to make sure they give identical results Batch job submission implemented and working PID Cuts, CT Cuts - Left/Right/Cent - Dummy + rand sub - Does order of cuts, diamond vs dummy/random matter? - Diamond cut is set based upon shape of low espilon diamond, it is defined by spectrometer acceptance, and will be the same for dummy/random - DG - Do diamond cut before the dummy/random subtraction since the cuts need to be applied on an event-by-event basis - Diamond cuts - > switch order of this, do this before dummy + rand - Phi/t binning -> also do this before the dummy/random sub - Get sample, then do dummy + random subtraction per phi/t bin - obtain yields per phi/t bin after applying all corrections and cuts - Ran all SIMC for all Q2 values for the kaon analysis - 5.5, 4.4 - With all of this info, *should* be able to run fortran script - TH: Q - Looking at CT spectrum, two random peaks near first central peak - RT: Seeing similar with other runs - TH: Q was what these were, was this resolved? - RT: Seems to be an issue with the right setting, some bug for this? - CT cut issue? - Right setting CT cuts don't seem to be getting applied correctly? Ali Updates ----------- Follow up from last week's slides re. cal and HMS-Cer efficiencies - Reviewed run sheets for all HeeP singles data in KaonLT run periods - Selected a few settings where momentum of HMS < pion threshold - Using these to get efficiency values - had some issues with replay script and needed to resubmit the jobs - Promises lots of plots next week Vijay Updates ------------- Checked t resolution using SIMC - Q2 = 0.38, low eps, central SHMS setting - How wide is second t bin? 0.004 -> ~4x larger than resolution (and this is probably the worst case) - t resolution does get worse with bin - GH: Should compute ratio of width of bin to resolution - i.e. how many sigma is each bin in width - a check for a high epsilon setting should also be done, since the beam energy is higher, the expected resolution will be worse t-/phi binning - 8 t bins - 16 phi bins, same as in Tanja's thesis - Looked at LTsep fortran scripts - DG: Make upper edge of MMPi peak cut bigger - 0.98 upper edge rather than 0.96, so as to be in the flatter part of the radiative tail, since there is no evidence of inelastic events in the data - Shouldn't make a huge difference in the yields, but this will reduce the cut sensitivity systematic error, which is desirable Jacob Updates ------------- - Ill for last week - Get well soon! Nathan Updates -------------- Follow up to LD- setting finding from last week, where the number of counts after RefCut optiization was lower than Online stats - Wrong NGC cut, fixed - ... but this didn't change anything! - GH: Should check online conditions to see if we're comparing things that are really the same #runs, etc - NH: it should be possible to reanalyze the data using online settings to check - one question last week was if runs were missing between online/offline - NH: checked this, and couldn't find anything wrong - definitely merits more investigation, since if the #events is really lower than online then there is something wrong with the RefCut optimization for this setting Junaid Updates -------------- - Submitted NP assignments, should be caught up by next week - Will look at calibrations shortly - Set up weekly JJN meetings similar to the RAV meetings - Wednesday afternoon - 14:00 Regina/15:00 Eastern Next Meeting ------------ Thur Feb 23 @ 17:00 Eastern/16:00 Regina/15:00 Mtn/14:00 Pacific - PionLT will go first