Tracking and Alignment in the CMS detector

Size: px

Start display at page:

Download "Tracking and Alignment in the CMS detector"

Meryl Lloyd
5 years ago
Views:

1 Tracking and Alignment in the CMS detector Frédéric Ronga (CERN PH-CMG) for the CMS collaboration 10th Topical Seminar on Innovative Particle and Radiation Detectors Siena, October Contents 1 Introduction 2 Tracking 3 Impact of misalignment 4 Alignment 5 Conclusion

2 Introduction The CMS detector F. Ronga (CERN-PH-CMG) IPRD06 1 / 16

3 Introduction External muon devices R (c m) MB MB MB 4 MB 3 DT eta = ME 1 RPC CSC Z (c m) ME 2 ME 3 ME Drift tubes (barrel) Cathode strip chambers (endcaps) Resistive plate chambers (trigger + redundancy) F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 2 / 16

4 Introduction Inner tracking devices 207 m 2 of Si sensors 10.6 million strips 65.9 million pixels Strips pitch µm resolution µm Pixels size 100(rφ) 150(z) µm 2 resolution µm F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 3 / 16

5 Tracking Tracker reconstruction 1 Trajectory seeding Seeded from hits in 2 pixel detector layers Can use primary vertex constraint Alternative (pixel-less) seeding under study 2 Trajectory building (pattern recognition) Kalman filter inside-out, layer to layer Includes energy loss and multiple scattering All (best) candidates grown up to outermost layer 3 Trajectory cleaning Remove mutually exclusive candidates (ambiguities) Resolved based on shared hits 4 Track fitting and smoothing Do Kalman fit of all hits for each candidate, inside-out (fitter) Redo fit in opposite direction, starting with fitter state (smoother) Muon devices reconstruction: cf. Martijn Mulders talk. F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 4 / 16

6 Tracking Tracker performance Global Efficiency µ, pt=1gev µ, pt=10gev µ, pt=100gev η Global Efficiency π, pt=1gev π, pt=10gev π, pt=100gev η 1 µ reconstruction efficiency 2 π reconstruction efficiency 3 transverse impact parameter resolution ) [µm] 0 d σ(δ 2 10 µ, pt=1gev µ, pt=10gev µ, pt=100gev ) [%] t /p t p σ(δ 10 µ, pt=1gev µ, pt=10gev µ, pt=100gev 4 transverse momentum resolution η η F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 5 / 16

Impact of misalignment Assessing the impact of misalignment Ideal detector geometry misaligned according to two scenarios : 1 First data scenario Situation at LHC start-up (first few 100 pb 1 );

7 Impact of misalignment Assessing the impact of misalignment Ideal detector geometry misaligned according to two scenarios : 1 First data scenario Situation at LHC start-up (first few 100 pb 1 ); information from construction data, laser; track-based alignment in Pixel detector. 2 Long term scenario Situation after a few fb 1 ; alignment at sensor level 20µm Alignment uncertainties in µm for the two scenarios (tracker). F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 6 / 16

Impact of misalignment Impact of misalignment on track reconstruction Global efficiency and transverse momentum resolution vs η for perfect alignment

8 Impact of misalignment Impact of misalignment on track reconstruction Global efficiency and transverse momentum resolution vs η for perfect alignment and the various scenarios Taking into account alignment uncertainty improves tracking efficiency. F. Ronga (CERN-PH-CMG) IPRD06 7 / 16

9 Alignment Alignment strategy Requirement: better than intrinsic resolution Tracker: determine 100k parameters at a precision 10µm Muons: 5k parameters at a precision of µm Concept in numbers Muon [µm] Tracker [µm] Strip Pixel Construction O(mm) Optical alignment 100 < 100 N/A Track-based 100 (or less) 10 5 Remarks Tracker: optical alignment ensures pattern recognition, track-based for final alignment (essential for pixels). Muons: optical alignment provides operational level, track-based alignment as cross-check and completion. F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 8 / 16

Alignment Construction knowledge Inputs Robots for module production Coordinate measurement machines Photogrammetry Provides initial position corrections; alignment position error.

10 Alignment Construction knowledge Inputs Robots for module production Coordinate measurement machines Photogrammetry Provides initial position corrections; alignment position error. Res: TIB-Lay1ext-Rod2-Mod Muon chamber X measurements h1 Entries 195 Mean RMS Res: TIB-Lay1ext-Rod2-Mod Res: TIB-Lay1ext-Rod3-Mod Tracker residuals without h3 Res: TIB-Lay1ext-Rod3-Mod2 and with survey h4 info. Entries 291 Mean RMS h2 Entries 141 Mean RMS Entries 214 Mean Preliminary 6.257e-05 RMS F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 9 / 16

Alignment Optical alignment overview Components Internal muon alignment barrel endcap Internal tracker alignment Muon w.r.t. tracker (Link system) Specifications Tracker structures 10µm Muon chambers at 100µm Muon vs tracker 100µm F.

11 Alignment Optical alignment overview Components Internal muon alignment barrel endcap Internal tracker alignment Muon w.r.t. tracker (Link system) Specifications Tracker structures 10µm Muon chambers at 100µm Muon vs tracker 100µm F. Ronga (CERN-PH-CMG) IPRD06 10 / 16

Alignment Optical alignment concept Barrel muon Monitor all chambers; complex triangulation.

r.t. TIB. Beams treated like tracks r/mm η 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.1 1.2 1.3 1.

5 1200 1100 TOB TEC 1000 900 BS ray 2 800 700 BS ray 3 600 500 ray AT 4 BS 400 AR 300 200 TIB

12 Alignment Optical alignment concept Barrel muon Monitor all chambers; complex triangulation. Endcap muon Monitor selected chambers (23% of all chambers) Tracker (laser beams) TEC w.r.t. TOB; TEC w.r.t. TIB. Beams treated like tracks r/mm η TOB TEC BS ray BS ray ray AT 4 BS 400 AR TIB TID ray z/mm F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 11 / 16

Alignment Track-based alignment overview Novel techniques developed to cope with large number of parameters Three different algorithms implemented: HIP algorithm

Cosmic muons Beam halo muons Muons from J/ψ and inclusive B decays High p t hadrons from QCD events (pixel) Combine with optical and survey data Take benefit

13 Alignment Track-based alignment overview Novel techniques developed to cope with large number of parameters Three different algorithms implemented: HIP algorithm (iterative method) Kalman filter algorithm (extension of track fitter) Millipede-II (full matrix calculation) Use of several data samples: Muons from Z µµ, W µν Cosmic muons Beam halo muons Muons from J/ψ and inclusive B decays High p t hadrons from QCD events (pixel) Combine with optical and survey data Take benefit mass and vertex constraints, overlap regions More details in the LHC Alignment Workshop (CERN, September 2006) F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 12 / 16

14 Alignment Track-based alignment study [tracker I] HIP algorithm example First data scenario Pixel barrel modules 200k Z µµ 10 iterations RMS 7µm in x,y RMS 23µm in z F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 13 / 16

Alignment Track-based alignment study [tracker II] Kalman filter 35k single µ p t = 100 GeV/c TIB layers 1 4 alignment RMS 21µm MillePede II TIB + TOB alignment 12015 alignment parameters

15 Alignment Track-based alignment study [tracker II] Kalman filter 35k single µ p t = 100 GeV/c TIB layers 1 4 alignment RMS 21µm MillePede II TIB + TOB alignment alignment parameters MillePede I: 13h MilliPede II: 32s (!) Residuals in local x vs. number of processed tracks Residuals in rφ before and after (RMS 4.8µm) F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 14 / 16

Alignment Track-based alignment study [muon] 790 chambers 5000 parameters Large amount of material Chambers considered as rigid body Two methods: tracks extrapolated from

16 Alignment Track-based alignment study [muon] 790 chambers 5000 parameters Large amount of material Chambers considered as rigid body Two methods: tracks extrapolated from tracker standalone muon tracks Alignment of barrel chambers with standalone muon tracks from W µν corresponding to 50h at F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 15 / 16

17 Conclusion Summary & Outlook Alignment of tracking devices in CMS is a challenge! Information from several sources (construction, optics, tracks) provide starting point and/or redundancy Complex track-based alignment algorithms developed use various data samples exploit physical/geometrical constraints use/combine with other sources Ongoing analysis of test beam and cosmic data valuable experience gained! Preliminary Proceeding well towards first physics data! F. Ronga (CERN-PH-CMG) Track&Align@CMS IPRD06 16 / 16

CMS Silicon Strip Tracker: Operation and Performance

CMS Silicon Strip Tracker: Operation and Performance Laura Borrello Purdue University, Indiana, USA on behalf of the CMS Collaboration Outline The CMS Silicon Strip Tracker (SST) SST performance during