|
1
|
Oct 5/22 PionLT/KaonLT Analysis Meeting Notes
|
|
2
|
---------------------------------------------
|
|
3
|
(Notes by GH and SJDK)
|
|
4
|
|
|
5
|
Today: KaonLT will be discussed first
|
|
6
|
|
|
7
|
Please remember to post your slides at:
|
|
8
|
https://redmine.jlab.org/projects/kltexp/wiki/Kaon_LT_Meetings
|
|
9
|
|
|
10
|
Present:
|
|
11
|
Regina - Garth Huber, Stephen Kay, Ali Usman, Love Preet, Alicia Postuma,
|
|
12
|
Vijay Kumar
|
|
13
|
CSULA - Konrad Aniol, Erika Gwin, Johnathan Conrad
|
|
14
|
Ohio - Jacob Murphy, Julie Roche
|
|
15
|
CUA - Richard Trotta, Tanja Horm
|
|
16
|
JLab - Dave Gaskell
|
|
17
|
FIU - Pete Markowitz
|
|
18
|
|
|
19
|
Richard Updates
|
|
20
|
---------------
|
|
21
|
- Bill's LT-separation code
|
|
22
|
- Two bugs to deal with, going to contact Bill soon
|
|
23
|
- C++ compiling issue
|
|
24
|
|
|
25
|
- Luminosity Updates
|
|
26
|
- Low beam on time runs
|
|
27
|
- Previously used BCM4A, used BCM1
|
|
28
|
- One variable index was set to 4A rather than 1
|
|
29
|
- Current cut applied on BCM4A - which was reading 0
|
|
30
|
- BCMs can be very different, 1 seems to work best
|
|
31
|
- I think this is the one Dave M recommended?
|
|
32
|
- Should use GoodStartTime in addition to GoodScinHit
|
|
33
|
- Tracking and PID cuts improved
|
|
34
|
- separate pi, p cuts for SHMS +polarity
|
|
35
|
- GH: could take all valid positive particles instead
|
|
36
|
- Things starting to look a lot better, particularly with the relative yield
|
|
37
|
without tracking
|
|
38
|
- Possibly correlated with weird behaviour of tracked yield?
|
|
39
|
- Finds that a discrepancy between EDTM-LT and CPU-LT is often due to bad
|
|
40
|
current cut
|
|
41
|
- 10.6GeV Lumi scan:
|
|
42
|
- SHMS relative scaler yield still looks odd
|
|
43
|
Carbon +14% slope @ 60uA
|
|
44
|
LH2 +30% slope
|
|
45
|
- SHMS untracked carbon looks very good except for 1 outlier
|
|
46
|
- SHMS tracked carbon still has 25% slope @ 70uA
|
|
47
|
- overall HMS scans looking good
|
|
48
|
- 8.2GeV Lumi scan (both spectrometers negative):
|
|
49
|
- change to BCM1 dramatically improves things, as most runs were
|
|
50
|
completely cut out when using BCM4A
|
|
51
|
- LH2: might actually see real -5% boiling, as trend is consistent across
|
|
52
|
scaler, untracked and tracked analyses, on both HMS, SHMS
|
|
53
|
- 6.2GeV Lumi scan:
|
|
54
|
- only 3 runs that look good (so far), up to 30uA
|
|
55
|
- this needs to be rechecked
|
|
56
|
- -ve polarity luminosities looking OK, converging
|
|
57
|
|
|
58
|
- HeeP Uncertainties
|
|
59
|
- Included uncertainties
|
|
60
|
- Stat uncertainty on Cherenkov?
|
|
61
|
- DG: Should be a global efficiency that doesn't vary run to run
|
|
62
|
- determine this via a dedicated Cherenkov efficiency study (HMS)
|
|
63
|
- Dave G
|
|
64
|
- Uncertainties probably should not just be sqrt(n)
|
|
65
|
- This is an overestimate
|
|
66
|
- Binomial errors instead
|
|
67
|
https://en.wikipedia.org/wiki/Binomial_distribution
|
|
68
|
- Cherenkov efficiency is like a Yes/no question each time
|
|
69
|
|
|
70
|
- To do
|
|
71
|
- Bill's cross section code
|
|
72
|
- Lumi Analysis
|
|
73
|
- Fast Raster Correction from FPi2
|
|
74
|
- *Might* already be done in hcana?
|
|
75
|
- In Fpi2 the calibration was off, can take a look
|
|
76
|
- Beam/target positions
|
|
77
|
- confirm dimensions compared to Dave Meekins write-up
|
|
78
|
- DG old document (2007) from XEM experiment
|
|
79
|
- DG: F2 people looked at this in Spring-18 data, need to locate this in
|
|
80
|
the hclog
|
|
81
|
- Calorimeter calibrations
|
|
82
|
- HGC efficiency calculation (Ali has write-up, need to add to code)
|
|
83
|
- Aerogel efficiency too
|
|
84
|
|
|
85
|
Discussion
|
|
86
|
----------
|
|
87
|
- SOS spectrometer acceptance cut in Data
|
|
88
|
- Mentioned in Blok paper
|
|
89
|
- Finds the boundaries of the acceptance
|
|
90
|
- Measured DIS scattering from deuterium, detailed comparison with SIMC
|
|
91
|
- SOS suffered significant saturation effects, which why it was needed
|
|
92
|
- this was not needed for HMS, should also not be needed for SHMS
|
|
93
|
|
|
94
|
- Richard mentions a tool to port in papers to a bibliography quickly
|
|
95
|
- https://www.zotero.org/
|
|
96
|
|
|
97
|
Vijay Update
|
|
98
|
------------
|
|
99
|
- Mainly writing committee report
|
|
100
|
- Updated shell script for lumi analysis, processed relevant lumi runs for low
|
|
101
|
energy data
|
|
102
|
- Need to process data through yield scripts
|
|
103
|
- Yield vs Current study is next after the report is done
|
|
104
|
|
|
105
|
Discussion
|
|
106
|
----------
|
|
107
|
- Error bars on the momentum offests
|
|
108
|
- Figure 4.7 (page 61) in Jochen Volmer's thesis -
|
|
109
|
https://misportal.jlab.org/ul/publications/view_pub.cfm?pub_id=5412
|
|
110
|
- How were the error bars here determined?
|
|
111
|
- From examining width of W peak?
|
|
112
|
- Yes, based on sensitivity from kinematics (Delta-W)*(dp/dW)
|
|
113
|
- seems to be a fixed +/-0.1% uncertainty for 5 highest points
|
|
114
|
- One point has very large error bar
|
|
115
|
- Vijay notes that this point also has an angle offset
|
|
116
|
- Magnets should really behave in a smooth and consistent way
|
|
117
|
- Expect a trend on both spectrometer momenta, not neccessarily a trend for
|
|
118
|
the angle (should be reproducible at the same angle)
|
|
119
|
- Generally have a single in-plane angle offset
|
|
120
|
- What about the vertical angle offset?
|
|
121
|
- Dave G doesn't recall seeing this for Hall C in the past
|
|
122
|
- Could SHMS have affected this? Heavier than the SOS, and Walter
|
|
123
|
says the hall floor has sagged
|
|
124
|
- Expect small point-to-point variation in the angle offset
|
|
125
|
- Start with one angle offset and one momentum offset in each spectrometer,
|
|
126
|
and see if you can get a global solution, allowing only a small
|
|
127
|
point-to-point variation
|
|
128
|
|
|
129
|
Ali Updates
|
|
130
|
-----------
|
|
131
|
- Also working on committee report
|
|
132
|
- BCM "calibrations" (the one that checks the beam on/off periods) for Lumi
|
|
133
|
- Double check whether this was done for the HeeP analysis
|
|
134
|
- Wasn't, added scaler replay and this determination back in
|
|
135
|
- Will check the uncertainty in the efficiencies, as per our discussion
|
|
136
|
|
|
137
|
Jacob Updates
|
|
138
|
-------------
|
|
139
|
- Chatting to Peter Bosted
|
|
140
|
- Posted something to SIDIS log
|
|
141
|
- He applied 6.2 GeV HMS matrix, sharpened peak and reduced MM tail
|
|
142
|
- should have beter agreement with MC peak shape now
|
|
143
|
|
|
144
|
Stephen Updates
|
|
145
|
---------------
|
|
146
|
- Discussion of John Matter's proton absorption spreadsheet
|
|
147
|
- Lots of the legwork is done, a few things we'll need to adjust
|
|
148
|
- probability of proton coming in, that gets through for trigger
|
|
149
|
- material thicknesses in g/cm2
|
|
150
|
- uses average of nuclear collision length (Lambda_T) and nuclear
|
|
151
|
interaction length (Lambda_I)
|
|
152
|
- probability goes as p(x)=exp(-Sum(x/Lambda)) for total absorption
|
|
153
|
- this is based on the idea that won't lose elastic scatt (forward) in
|
|
154
|
trigger, but can lose multiple elastic scatterings
|
|
155
|
- calculation is for 2.25 hodo planes, corresponding to minimal 3/4 trigger
|
|
156
|
|
|
157
|
- Switch proton Lambda values to pion/kaon as appropriate
|
|
158
|
- Kaon values not directly available
|
|
159
|
- Scale pion values by difference in cross section at sqrt(s) for our
|
|
160
|
reaction?
|
|
161
|
- Need to make some reasonable assumption that we can justify and use
|
|
162
|
this
|
|
163
|
- plots of pi/K cross sections p.782 of current version of PDG
|
|
164
|
|
|
165
|
- Need to switch HGC gas from CO2 to C4F10 as actually used in our expt
|
|
166
|
- Remove NGC and replace with scattering chamber as needed
|
|
167
|
- Adjust aerogel based upon refractive index used, i.e. density scales as
|
|
168
|
(n-1)
|
|
169
|
- Adjust LH2 thickness depending upon spectrometer angle
|
|
170
|
|
|
171
|
Next meeting: Wed Oct 12 @ 9:00 Eastern/7:00 Regina/6:00 Calif
|
|
172
|
- PionLT will start first
|
|
173
|
|
|
174
|
|