calibration ANTARES-KM3NeT Coll. Meeting Granada, 07/05/2018 Nafis - - PowerPoint PPT Presentation

▶
calibration
SMART_READER_LITE
LIVE PREVIEW

calibration ANTARES-KM3NeT Coll. Meeting Granada, 07/05/2018 Nafis - - PowerPoint PPT Presentation

Mar 10, 2023 269 likes •415 views

Last news on muon time calibration ANTARES-KM3NeT Coll. Meeting Granada, 07/05/2018 Nafis Rezwan Khan Chowdhury Ankur Sharma IFIC INFN Sezione di Pisa Agustn Snchez Losa INFN Sezione di Bari Since Rabat Studies of

slide-1
SLIDE 1

Agustín Sánchez Losa INFN – Sezione di Bari

Last news on muon time calibration

ANTARES-KM3NeT Coll. Meeting

Granada, 07/05/2018

Nafis Rezwan Khan Chowdhury IFIC Ankur Sharma INFN – Sezione di Pisa

slide-2
SLIDE 2

Since Rabat…

  • Studies of correction/offset evolution per iteration
  • Fitting offsets and control plots
  • Update of the two offsets computed so far (ARCA/ORCA)
  • Comparison with available offsets
  • Preliminary findings on MC
  • Creating a centred documentation of the muon calibration

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 2

slide-3
SLIDE 3

Muon time calibration

  • Determination of detector time constants using atmospheric muon

tracks as time reference between detector elements

  • Only method there where optical beacons are not available…
  • ...but not a prompt tool: requires many runs for statistics
  • Recipe:

1) reconstruct tracks excluding the component to calibrate (DOM, DU) 2) collect hit time residuals (HTR) of the tracks with the component to calibrate 3) compare with the rest of the detector HTR to estimate relative corrections 4) include found corrections on the calibration and repeat until reach stability

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 3

slide-4
SLIDE 4

Program flow

  • Raw DATA method:
  • DATA KM3NeT runs
  • Event selection
  • Loop on find corrections:
  • JPreSim_HTR
  • Offsets per iteration
  • Final offsets
  • MC vs DATA method:
  • DATA & MC KM3NeT runs
  • Event selection
  • Loop on find corrections:
  • JPreSim_HTR
  • Jfit_HTR
  • Offsets per iteration
  • Final offsets

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 4

JTriggerProcessor / JTriggerReprocessor

Event selection: 7 L1 DOM hit

JPreSim_HTR

JPreSim+JPreFit track reconstruction + HTR

JFit_HTR

JFit DATA-MC HTR comparison

DATA MC

KM3NeT_DETID_RUN.root

HTR HTR Offsets Offsets

KM3NeT_-DETID_RUN.root

slide-5
SLIDE 5

Studies on evolution per iteration

Corrections found per iteration:

  • Global offset of HTR median at

~1.7-2.0 ns (DU dependent)

  • Main transitory phase (non-stable

region): biggest corrections

  • Fine tuning phase (stable region):

asymptotic small corrections How many iterations required?

  • Depends how the offset is computed

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 5

Evolution of corrections found per iteration for ARCA_00005009_00005046 DATA

slide-6
SLIDE 6

Fitting offsets

  • InterDOM offsets estimation from iteration plots:
  • A) Average value of last 10 iterations:
  • Simple to perform and reproduce
  • Different asymptotic behaviour not into account
  • B) Fitting to a model:
  • Complex to define an stable fit
  • Asymptotic behaviour into account on offset and error estimations
  • Chosen fit:
  • Simultaneous fit on all the available DOMs plus σ𝑗

𝐸𝑃𝑁𝑡 𝑃𝑔𝑔𝑡𝑓𝑢𝑗 = 0 per iteration

constrain

  • Fit: 𝑧𝐸𝑃𝑁 it = 𝑃𝑔𝑔𝑡𝑓𝑢𝐸𝑃𝑁 + 𝛽𝐸𝑃𝑁

1

∙ 𝑓−𝜇1∙ it + 𝛽𝐸𝑃𝑁

2

∙ 𝑓−𝜇2∙ it

  • A) – B) differences <0.02 ns on stable DOMs but up to ~0.2ns on difficult

cases

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 6

slide-7
SLIDE 7

Fitting offsets

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 7

Asymptotic exponential fitting on ARCA_00005009_00005046 DATA

slide-8
SLIDE 8

InterDOM offsets

#DOM DU001 DU002 1

+8.0 +11.1

2

+8.3

  • 3

+7.5 +12.2

4

+6.5

  • 1.4

5

+5.3 +7.7

6

+3.1 +11.0

7

+1.6 +1.6

8

+1.6 +7.1

9

  • 1.3
  • 2.4

10

  • 0.9
  • 5.2

11

  • 3.3
  • 12
  • 1.2
  • 13
  • 7.5
  • 11.2

14

  • 4.8
  • 3.0

15

  • 6.7
  • 5.3

16

  • 7.5
  • 9.3

17

  • 8.8
  • 3.1

18

  • 9.7

OFFSETS (ns)

Computed for ARCA

  • Code name: ARCA_00005009_00005046
  • 38 runs (5009-5046)
  • 13/12/2016 - 01/01/2017
  • DETX: all_DB_PMT_noref.detx

Karel’s detector file

  • t0: ?
  • geo: ?
  • w/o MC + JTP
  • 100 iterations (asymptotic fitting)

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 8

Next slide: comparison with Karel’s (w MC and same detx)

slide-9
SLIDE 9
  • 15.0
  • 10.0
  • 5.0

+0.0 +5.0 +10.0 +15.0 1 2 3 4 5 6 7 8 9 101112131415161718 InterDOM offset (ns) ARCA DU002 DOM#

  • 15.0
  • 10.0
  • 5.0

+0.0 +5.0 +10.0 +15.0 1 2 3 4 5 6 7 8 9 101112131415161718 InterDOM offset (ns) ARCA DU001 DOM# Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 9

# Offset (ns) w/o MC w MC DIFF 1

+8.0 +5.4

+2.6 2

+8.3 +5.7

+2.5 3

+7.5 +5.2

+2.3 4

+6.5 +4.9

+1.6 5

+5.3 +4.0

+1.3 6

+3.1 +2.7

+0.4 7

+1.6 +1.6

  • 0.0

8

+1.6 +2.1

  • 0.5

9

  • 1.3
  • 0.3
  • 1.0

10

  • 0.9

+0.2

  • 1.0

11

  • 3.3
  • 2.1
  • 1.2

12

  • 1.2

+0.1

  • 1.3

13

  • 7.5
  • 6.2
  • 1.3

14

  • 4.8
  • 3.4
  • 1.4

15

  • 6.7
  • 5.5
  • 1.1

16

  • 7.5
  • 6.7
  • 0.8

17

  • 8.8
  • 7.5
  • 1.3

18

  • ARCA-DU1

# Offset (ns) w/o MC w MC DIFF 1

+11.1 +8.3

+2.8 2

  • 3

+12.2 +10.0

+2.2 4

  • 1.4
  • 3.3

+1.9 5

+7.7 +6.3

+1.4 6

+11.0 +9.8

+1.2 7

+1.6 +6.0

  • 4.4

8

+7.1 +6.6

+0.5 9

  • 2.4
  • 2.6

+0.2 10

  • 5.2
  • 5.1
  • 0.1

11

  • 12
  • 13
  • 11.2
  • 10.6
  • 0.6

14

  • 3.0
  • 2.2
  • 0.9

15

  • 5.3
  • 4.2
  • 1.1

16

  • 9.3
  • 8.0
  • 1.3

17

  • 3.1
  • 2.1
  • 1.0

18

  • 9.7
  • 9.0
  • 0.7

ARCA-DU2

  • General good agreement
  • Two main differences:
  • At bottom DOM → MC?
  • DU002 DOM07 → Karel’s step fluctuations during

iterations?

slide-10
SLIDE 10

InterDOM offsets

#DOM DU002 1

+4.5

2

+4.1

3

+4.3

4

+4.4

5

+3.5

6

+3.3

7

+1.9

8

+1.4

9

+1.1

10

+0.5

11

  • 0.4

12

  • 0.7

13

  • 2.2

14

  • 3.5

15

  • 4.1

16

  • 5.0

17

  • 6.7

18

  • 6.1

OFFSETS (ns)

Computed for ORCA

  • Code name: ORCA_00003165_00003192
  • runs (3165-3192)
  • 04 - 08/12/2017
  • DETX: KM3NeT_00000029_20170920.detx

Same used by Rodrigo for NBs

  • t0: ?
  • geo: ?
  • w/o MC + JTR
  • 100 iterations (average last 10)

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 10

Next slide: comparison with Rodrigo’s (NB and same detx)

slide-11
SLIDE 11

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 11

# Offset (ns) w/o MC NB DIFF 1

+4.5 +12.7

  • 8.2

2

+4.1 +10.6

  • 6.5

3

+4.3 +8.4

  • 4.1

4

+4.4 +7.5

  • 3.1

5

+3.5 +5.0

  • 1.5

6

+3.3 +4.0

  • 0.7

7

+1.9 +3.0

  • 1.1

8

+1.4 +1.4 +0.0

9

+1.1

  • 0.3

+1.4

10

+0.5

  • 1.9

+2.4

11

  • 0.4
  • 3.4

+3.0

12

  • 0.7
  • 3.7

+3.0

13

  • 2.2
  • 4.1

+1.9

14

  • 3.5
  • 5.2

+1.7

15

  • 4.1
  • 6.9

+2.8

16

  • 5.0
  • 7.7

+2.7

17

  • 6.7
  • 10.0

+3.3

18

  • 6.1
  • 9.3

+3.2

ORCA-DU2

  • 15.0
  • 10.0
  • 5.0

+0.0 +5.0 +10.0 +15.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 InterDOM offset (ns) ORCA DU002 DOM#

A preliminary interpretation of NB (N)-(N+1) Δt into offsets:

  • NBs seems much more sensitive than Muon
  • Muon method may suffer of some bias on lower DOMs, stronger if MC

comparison is used (see ARCA DU1)

  • Both methods suggest the possibility of an incorrect interDOM distance:

time/distance slope compatible with a ~3% DU elongation (also on ARCA)

  • If this is correct:
  • previous offset plot slope tendencies are explained
  • …yet ARCA DU2 features are not explained
slide-12
SLIDE 12

Preliminary findings

  • n MC
  • InterDOM offsets look on

ARCA_00005009_00005097 MC:

  • MC:

/hpss/in2p3.fr/group/km3net/mc/KM3NeT_ 00000014/Runs_5009- 5097/JTE/v3.1/KM3NeT_- 00000014_02072016_mu10G_*.km3Hit2_v5r 4.JTERun50*Eff05.root

  • detx:

/afs/in2p3.fr/throng/km3net/detectors/KM 3NeT_-00000014_02072016.detx

  • No transitory phase (only fine tuning)
  • Some systematic bias starts to be

confirmed… active DOMs dependant?

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 12

slide-13
SLIDE 13

Documentation and official offsets

  • Building a kind of “storefront” of the muon calibration at GIT:
  • http://git.km3net.de/muon_calibration

./documentation [→ really incomplete at this stage: first step is add all talks] ./code [→ really incomplete at this stage] ./calibrations

./interDOMoffsets [→ ready for the only offsets done so far] ./interlineoffsets [→ for the future]

  • Easy to communicate to other calibration groups and to the collaboration
  • Aim to stablish a common language:
  • in order to compare between different calibrations clearly and less prone to

errors: offset format proposal, used runs, methods…

  • Difficulties in defining unequivocally conditions of a certain calibration:
  • Assumed geometry → DB table reference? Real-time geometry version + runs?
  • Assumed t0 initial calibration → t0 table used to build the .detx file? Info at DB?

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 13

slide-14
SLIDE 14

Conclusions

  • Studies on the evolution per iteration: understanding the method
  • Fitting procedure: more control on offset and error estimations
  • General agreement between methods: similar things are seen (small

differences still to be understood)

  • Systematic bias evidence in MC: parametrisation required
  • Incoming plan to study effects with MC: active DOM dependence, etc
  • Call for coordination and common language-communication

standards among time calibration groups

Granada, 07/05/2018 Agustín Sánchez Losa – INFN Sezione di Bari | ANTARES-KM3NeT Coll. Meeting 14