Feb 29/24 PionLT/KaonLT Analysis Meeting Notes
----------------------------------------------
(Notes by GH)
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 - Nathan Heinrich, Muhammad Junaid, Ali Usman, Garth Huber, Vijay Kumar,
Nacer Hamdi, Alicia Postuma
CUA - Richard Trotta, Casey Morean
JLab - Dave Gaskell
FIU - Pete Markowitz
Richard
-------
- PhD thesis has now been submitted to the committee for review.
Congratulations on this milestone!
LT-Iteration Script Issues
- uncertainties for unseparated d2sigma behave weirdly when iterating, while
the R=model/data ratios behave normally
- statistical errors on d2sigma get much smaller after an iteration
- clearly there is still a bug somewhere
- GH: it sounds like the way the R-error is propagated to d2sig-error is the
issue
Plots of d2sigma vs phi are shown for various settings at high, low epsilon
- a MM cut is now applied: 1.1high epsilon issue
- clearly this will need significantly more investigation
Ali
---
- was sick last week, only getting back on track now
Clean pi+,p data for Richard's PID leakage subtractions
- finished analyzing pi+ for 3 of 5 kinematics, data ready for RT subtraction
- proton will be done soon
- no HGC cut used at all
- RT will apply the HGC hole cut to his sample, confirms that it will be
straightforward to implement
- RT still has to resolve the 5cm HGC hole cut shift issue reported a few
weeks ago
- RT thinks the x_HGC, y_HGC coordinates were not implemented correctly,
will return to this soon
Resuming SIMC resolution study, nothing to report yet
Alicia
------
HMS-delta correction update
- had a discussion with DaveG a few days ago, after finding that application of
DG's delta-correction did not yield the expected improvement to MM resolution
- found that DG's MM calc was based on the old 6GeV/ENGINE formula
- Alicia is using hcana calculation, and this appears to be the cause of the
difference
- DG is trying to figure out what exactly is different between the two MM
calculations
- hcana: takes x',y' and spectrometer angles, converts to spherical theta,phi
and calculates PM components
- ENGINE: calculates PM components in spectrometer x',y' system and converts
to target coordinates, but does not use spherical coordinates at any step
- DG: calc in hcana is harder to follow, can't see where the difference is,
in principle both calculations should be equivalent
- AP: tried to recalculate MM on her own, also agreed that old method is
easier to understand
- her calculation does not fully reproduce hcana either
- Alicia shows a plot from script, sees no correlation between delta and hsxpfp
for P_HMS<5 GeV/c
- Q2=2.1, W=2.95, P_HMS=5.292: starts to see a correlation
- Q2=3.0, W=2.32, P_HMS=6.59: a definite curved correlation is seen
- will have to apply a correction for these two settings, but it needs to be
different than what DG had found
- DG: for P_HMS=6.59, are we 100% sure we're using the correct Matrix
Elements for this setting?
- Ali goes to check, confirms that the correct ME are used for both his and
Alicia's analyses
- Jacob was looking at finalizing the high HMS momenta ME, Holly is looking
at higher P_H MS w/NPS students
- clearly there is an issue that needs further optimization
- Implementing the HMS-delta correction for high P_HMS
- DG: can't just correct MM, all kinematics are affected, in principle will
affect the t-binning for these settings, as events will migrate from one
bin to another
- rather than Alicia continuing with her reconstruction script, it will be
better to modify hcana to include the correction for high P_HMS, so that
all kinematics are handled consistently
- DG: suggests fitting a quadratic, but only apply to a certain region of the
HMS focal plane
- Alicia: maybe a piece-wise linear function? Will think about what works
best
- *CONCLUSION* Alicia will first determine the needed delta correction then
implement in hcana
Vijay
-----
- Just returned from EIC workshop in India, starting analysis again and yield
ratios
Shows interesting Lattice QCD calculation from workshop
- Swagato Mukherjee (BNL), Q.Shi will be first author, not yet submitted
- LQCD calc for both pi+ and K+ to Q2~35
- calc indicates that both Q2*Fpi and Q2*Fk will remain high through the full
Q2 range, with no significant decrease to pQCD curve
- not surprisingly, the Lattice errors are fairly large
- we definitely need to keep a look out for the final calculation
Nathan
------
Working on PionLT Lumi study and plots for the SHMS NIM paper
- looking at what PID cuts to apply for the Lumi study, which also generates
the needed plots for the paper
- SHMS single arm e' Lumi scan
- NGC vs Calorimeter
- NGC vs Aerogel
- NGC vs HGC
- GH: suggests to apply NGC vs Calor cut and look at the others with this cut
applied to see what is removed before deciding whether the other plots add
anything extra to the PID selection (useful also for the paper)
- HGC, NGC efficiencies vs delta
- efficiency plots vs X_cer still in progress
- HGC, NGC, Aerogel 2D efficiency plots vs X,Y_FP
- see hole in the HGC, NGC and tray boundaries in Aerogel
- LH+ production data PID plots shown
- LD- also quickly looked at, sees large e- peak in SHMS
Next Steps:
- make sure sure Lumi script is working properly
- RT will meet with Nathan separately to go over what changes he needs to
make to Lumi script
- In addition to changes since 2021, Nathan will have to modify the script to
handle Lumi runs where only one arm data at a time (should be straightforward)
- apply proper cuts on data ranges
Junaid
------
Working on PionLT efficiency studies
- investigating outlier runs, removed bad runs from plots
- most of these had beam trip issues
- also some runs with quadrupole trips
- DG: how do you remove these runs from the analysis?
- run list is modified, so these runs will never be replayed in future jobs
- tracking efficiency studies nearly complete
- generating beam cut files from replays
- will be included in future replays
Heep-check analysis preliminaries
- looked closely at the kinematics for all Heep-Coin settings in 2021-22
- selected 9 of the 16 settings for further study, based on the angle and
momentum settings
- GH: the other 7 settings will be kept in reserve for a future student to
look at
- did not look at list of Heep-singles runs yet, it's a very large list of
settings
- will first look at the Heep-coin results and from there decide which
Heep-singles settings are helpful to look at
Nacer
-----
Efficiency studies for 3.8, 4.9 GeV KaonLT data
- Plot of CPULT vs COIN trigger rate
- sharp drop off seen above 2.8 kHz
- looked at the outlier run near 2.8 kHz, where the CPULT relation kink
- no comment in run sheet, will look next at hclog
- DG: are the rates corrected for the beam off time?
- will have to look more closely into this
- Plot of SHMS_pion_sing_TrkEff
- vs SHMS_3/4 trigger rate
- 2 outliers seen
- vs S1X rate
- 3 outliers seen
Next Meeting
------------
- Thur Mar 7 @ 15:00 Eastern/14:00 Regina
- PionLT will go first
- USA daylight savings switch is following week
- meeting time will remain at 15:00 Eastern, which switches to 13:00 Regina
- GH will send Zoom invitations up to end of August