## Kaon LT Meetings » mtg_23aug17.txt

1 |
Aug 17/23 PionLT/KaonLT Analysis Meeting Notes |
---|---|

2 |
---------------------------------------------- |

3 |
(Notes by GH) |

4 | |

5 |
Today: PionLT 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 |
------- |

12 |
Regina - Garth Huber, Nathan Heinrich, Nacer Hamdi, Alicia Postuma, |

13 |
Muhammad Junaid, Vijay Kumar |

14 |
JLab - Dave Gaskell |

15 |
CUA - Richard Trotta |

16 |
CSULA - Konrad Aniol |

17 | |

18 |
Nathan Updates |

19 |
-------------- |

20 |
HGCer Calib Update |

21 |
- parameter equation in standard.database |

22 |
- changes pushed to GitHub |

23 |
- comparison w/ online HGC plots |

24 |
- NPE distribution goes much higher now, particularly for last runs of |

25 |
experiment, where gain drop was significant |

26 |
- will post slides on calib after final set of runs is completed on ifarm |

27 | |

28 |
Working on standard.kinematics |

29 | |

30 |
Junaid Updates |

31 |
-------------- |

32 |
HMS Calorimeter Calib |

33 |
- working on 9.177 GeV (2021) data |

34 |
- marged Nathan's changes |

35 |
- calibrating once a shift now |

36 |
- DG: question on whether it is needed to calibrate this frequently |

37 |
- MJ: sees shifts after 2-3 days, so need fairly often |

38 | |

39 |
GH: please have a discussion with Nathan, and come up with a plan on how many |

40 |
runs need to be calibrated, and how long do you expect it to take |

41 |
- concerned that MJ will feel stuck working on calibrations for a very long |

42 |
time, so we should see if we can finish by Labor Day, if possible |

43 | |

44 |
Richard Updates |

45 |
--------------- |

46 |
Changes to mc_shms_hut.f for calorimeter edge cut |

47 |
- it was commented out in RT's version, but NOT commented out in SIMC master |

48 |
- DG: these cuts can be done either in a standalone script, or in SIMC, it's a |

49 |
matter of what you prefer |

50 | |

51 |
Aerogel tray cuts now implemented in recon_hcana script |

52 | |

53 |
Implementing HGC hole cut |

54 |
- will discuss this more w/ Ali after he's back |

55 | |

56 |
Kinematics averaging code is now simpler, replicating old Fortran code |

57 |
- now just taking average kinematics per t-bin |

58 | |

59 |
Will give an update on Lumi scans status next week |

60 | |

61 |
Vijay Updates |

62 |
------------- |

63 |
Confirmed that calorimeter edge cut is NOT commented out in his SIMC |

64 | |

65 |
Target boiling corrections work in progress |

66 |
- hoping to show some plots next week |

67 | |

68 |
Alicia Updates |

69 |
-------------- |

70 |
Follow up from Portia's work on piDelta BSA |

71 |
- Portia did not have weighted SIMC plots, adding weights improved things a lot |

72 |
- also a small error in phi conversion between Data & SIMC corrected |

73 |
- No Dummy subtractions yet (but Randoms are subtracted) |

74 | |

75 |
- Using SIDIS & piDelta MC output for MM-distribution shape |

76 |
- fit to data looks very good (Q2=2.115) |

77 |
- pi+n radiative tail also not yet included |

78 |
- plot including radiative tail shows a significant contribution underneath |

79 |
piDelta peak |

80 | |

81 |
- Preliminary BSA plot (Center only) |

82 |
- integrated fitted piDelta MC to data |

83 |
- expecting a large systematic uncertainty from the fit |

84 |
- only one t-bin so far, possibly the Q2=2.115 can have two t-bins |

85 | |

86 |
- DG: HERMES had only one t-bin, but they removed the -t_min region (where |

87 |
asymmetry is small) so that the remaining region had a larger BSA |

88 | |

89 |
pi+n Binning |

90 |
- Q2=5.5: decided to have 4 t-bins rather than 5 |

91 |
- trying different phi-bin options: 9, 11, 13 bins |

92 |
- DG: should get same asymmetry for all choices of phi-bins, provided there |

93 |
are sufficient bins to fit the asymmetry moments |

94 |
- in general, more bins are better |

95 | |

96 |
- 13 phi-bins looks good for Q2=5.5 |

97 |
- GH: please make a table of fitting results & errors vs #phi bins, so we can |

98 |
check fit stability and trends |

99 |
- DG: a little worried about the dependence on #phi bins, suggests a MC study |

100 |
to see if there are any issues, such as e.g. if 9 phi-bins are sufficient |

101 | |

102 |
- Q2=4.4: 5 t-bins seems fine |

103 |
- results with both 13, 15 phi-bins look okay |

104 |
- 13 bins has a bit more wiggles |

105 | |

106 |
- Q2=3, W=2.32: 7 t-bins |

107 | |

108 |
- Q2=3, W=3.14: 7 t-bins |

109 |
- this set has very good statistics |

110 | |

111 |
- Q2=2.115: 8 t-bins |

112 |
- first t-bin has nearly zero asymmetry, very close to -t_min |

113 |
- fit with A,B,C a bit weird, looks reminiscent of a capacitor breakdown |

114 |
plot |

115 |
- NG: there should be a way to tell ROOT the bins have a width, that |

116 |
might get rid of the weird fits |

117 |
- Nacer: has a small function that AP can try |

118 | |

119 |
BSA Error Analysis |

120 |
- tried to re-derive error equation in S.Diehl's paper |

121 |
- Diehl's eqn seems to be based on the approximation |

122 |
Y+ \sim Y- which leads to Y+^2 + Y-^2 \sim 2Y+Y- |

123 |
- DG confirms this is a common choice in PV experiments |

124 | |

125 |
- Alicia's formula has no approximations, is exact |

126 |
- approximate formula leads to an under-estimation of errors |

127 |
- difference between exact formula and approximate formula is small in most |

128 |
cases, but the deviation grows for points where the BSA > 0.1 |

129 |
- Diehl BSA < 0.1, so difference not significant for those data |

130 | |

131 |
- we will use Alicia's formula, since it is more accurate |

132 |
Next Meeting |

133 |
------------ |

134 |
- Thur Aug 24 @ 16:00 Eastern/14:00 Regina/13:00 Pacific |

135 |
- KaonLT will go first |

136 | |

137 |