Tracker Resolution/Update

Tracker

Resolution/Update

Melissa Uchida

Imperial College London

Analysis Parallel 30/3/16

CM44 30/3/16 2

Summary

●

Hardware Summary

●

Reconstruction Data

●

Software and Improvements

●

Efficiency

●

Resolution For Step IV

CM44 30/3/16 3

Hardware Summary

● The hardware has been running well since October 2015,

● One waveguide was swapped on 24^th Feb, this was implemented in mapping and reconstruction within 36 hours.

● Several re-calibrations have been performed (as

necessary) and calibration data is taken once per shift.

● Helium has been in both trackers since February and is operating well. Epics monitoring is underway.

● The hall probes have operated well throughout and produced useful data. They're currently unplugged

awaiting helium leak proofing ~2 weeks (before field on data/training).

CM44 30/3/16 4

Tracker Reconstruction and Data Validation

●

Data is analysed for validity within ~24 hours turnaround.

●

All reconstruction level plots and efficiency are available online through

http://www.hep.ph.ic.ac.uk/~mgeorge/Files/2016M ICEData/

●

There are two scripts available through MAUS for

validation and efficiency work and the full set and

structure will be implemented in a few weeks.

CM44 30/3/16 5

Digits, MAUS v2, Run7769 US

CM44 30/3/16 6

Digits, MAUS v2, Run7769 US

Data

Noise from calibration

CM44 30/3/16 7

Data

Noise

CM44 30/3/16 8

Digits, MAUS v2, Run7769 DS

CM44 30/3/16 9

Digits, MAUS v2, Run7769 DS

Data

Noise from calibration

LED calibration

CM44 30/3/16 10

Clusters, 7850

Ratio = 10 as Reqd

-- singlet -- doublet

CM44 30/3/16 11

Kuno Plots MAUS 2.1

Entries 320+ Central fibre offset by 1 – known bug

CM44 30/3/16 12

Spacepoints, 7850

CM44 30/3/16 13

Spacepoints, 7850

CM44 30/3/16 14

Beam Profiles US

7864

CM44 30/3/16 15

Beam Profiles DS

7864

CM44 30/3/16 16

Tracks

CM44 30/3/16 17

Track Finding Efficiency (Straights)

Upstream x residuals, Downstream y residuals.

Comparison with projected hit position from joining a hit in TOF1 and TOF2

Upstream x-y residuals

CM44 30/3/16 18

Track Finding Efficiency

If you have 5 spacepoints in tracker what is the efficiency of finding a 3-5 sp track, req a sp in TOF 1 and 2

US 100%

DS 99.85%

Binomial error

CM44 30/3/16 19

Track Finding Efficiency

0.84 0.85 0.86 0.87 0.88 0.89 0.9 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1 1.01

Pat Rec Track Finding Efficiency

TkUS_5pt TkUS_3-5_pt TkDS_5pt TkDS_3-5pt

Run

Efficiency

If you have 5 spacepoints in tracker what is the efficiency of finding a 3-5 or 5 sp track, req a sp in TOF 1 and 2

Binomial error

CM44 30/3/16 20

Tracker Efficiency from npe

● Unbiased by noise and does not require track finding.

– Uses cluster light yield to determine # noisy hits which contaminate SP doublets.

● Coincidence (20-50ns) between TOF1&2 trigger SP (within 2ns of particle trigger) and removing multiple events per trigger.

● SP grouped into doublets and triplets and light yield of clusters forming the SPs are stored.

● These are combined to find efficiency:

E =(T + D −N_D)/C

where T= triplets, D = doublets, ND = doublets from noise and C = # TOF coincidence.

● T, D and C taken from event selection

CM44 30/3/16 21

Tracker Efficiency

●

Prob of doublet sp from noise, P

≈ 3.6 × 10

●

N

determined from doublet light yield histogram fitted with triplet light yield and expon decay

(noise)

L

(x) = f

L

(x) + f

exp(ax)

x = pe, L

(x) = doublet light yield, L

(x)= triplet

light yield, f

and f

fits triplet histo and noise

height respectively

CM44 30/3/16 22

Efficiency Per Stn

– total – triplets

● Requires: TOF1-2: 0-100ns, TOF1 slabs 2,3,4, TOF2 slabs 4,5,6

● US eff: a track from the DS tracked to the active area of station 1 in the US.

● DS eff a stack from US tracked to the active area of station 1 in the DS

Melissa Uchida CM44 30/3/16 23

Tracker since CM43

● Efficiency measured at spacepoint level for each station.

● Correction applied to stop noise hits improving efficiency

● Requires:

– TOF1-2: 0-100ns

– TOF1 slabs 2,3,4

– TOF2 slabs 4,5,6

● For upstream efficiency a track from the downstream tracked to the active area of station 1 in the upstream.

● For downstream efficiency a stack from upstream tracked to the active area of station 1 in the downstream

CM44 30/3/16 24

Helical Track Finding

● Analysis of tracker resolution and efficiency for helical track

reconstruction is underway and will be published in the scifi paper soon.

● All plots are for a fairly idealised tracker simulation.

US pz resolution pz

DS pt resolution pt Chris Hunt &

Adam Dobbs

CM44 30/3/16 25

Helical Track Finding

US pz track efficiency

DS pt Track efficiency Chris Hunt &

Adam Dobbs

Melissa Uchida CM44 30/3/16

Step IV Fields in Reconstruction

●Ongoing studies of the effect of the non-uniform tracker field.

●Field Misalignment and Field Uniformity are being treated separately

●First results of Field Misalignment are being processed

● Simulate a tracker in perfect field

● Incrementally rotate the field a perform a full resolution study.

● Plot the changes in resolution as a function of field rotation.

●Field uniformity study is currently being conducted for the emittance paper

● Perform a full resolution study for perfect fields, opera field maps and MAUS field maps.

● Try to understand the difference in performance and how it depends on the field uniformity.

● Extend with different field maps and analyse the trackfit pulls to try to learn something.

Currently this study is being performed for use in the emittance paper.

Unfortunately it is not quite at the plot-producing phase.

Chris Hunt

CM44 30/3/16 27

Final Summary

●

Tracker Hardware is performing well and any

problems have been quickly resolved. Helium is in both trackers, hall probes are operating well

(currently unplugged awaiting helium leak proofing

~2 weeks).

●

Calibration noise and LED data is taken daily and analysed almost immediately, new calibrations are produced with fast turnaround.

●

Data taking is going well and data is analysed for validity/problems etc within 24 hours.

●

Efficiency is calculated in three ways, track finding efficiency, doublets vs triplets and npe and is

~99.5%

CM44 30/3/16 28

Digits MAUS v2 Helicals 7469

CM44 30/3/16 29