The Atlas combined testbeam MDTRPC BOS Tilecal PixelsSCT
The Atlas combined testbeam MDT-RPC BOS Tilecal Pixels&SCT TRT LAr Thijs Cornelissen, 1 NIKHEF Jamboree, Nijmegen, December 2004
2 Introduction n n Combined Testbeam: ~ March – November 2004 Full ATLAS barrel slice, starting from the interaction point (origin of the global coordinate system, same for all the detectors) All relevant sub-detectors participate Detectors in their real positions as in ATLAS Primary objectives & motivations: ¡ ¡ ¡ Combined data taking with all the detectors Testing of the whole Dataflow system (common LVL 1 trigger, EF, ATLAS Event Format, etc. ) Final prototypes of the electronics (RODs) Combined calorimetry studies Combined studies of the Inner. Detector Combined studies ID + Muons Noemen: H 8 Letertype Jargon! Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
3 Inner. Detector: Pixels: 6 modules, inside magnet (B=1. 4 T) SCT: 8 modules, also inside magnet TRT: 1/16 th of the barrel Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
4 Combined setup n Inner. Detector: ¡ ¡ ¡ n Calorimetry: ¡ ¡ n Pixels: 6 modules, inside magnet (B=1. 4 T) SCT: 8 modules, also inside magnet TRT: 1/16 th of the barrel LAr: barrel module 0 + cryostat Tile. Cal: 3 layers of modules Muon detectors: ¡ ¡ 1 barrel projective tower (MDTs + RPCs) 1 endcap sector (MDTs+TGCs) Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
5 Offline software flow Simulation and Detector response G 4 Sim CTB n Bytestream Data Preparation Digitization Raw Data Objects Prep. Raw. Data (Clusters and Drift circles) and Space Points The full ATLAS offline software is used for the testbeam data. Excellent test for various components: ¡ ¡ ¡ n Geant 4 Hits Bytestream converters Calibration and alignment Track reconstruction Some testbeam-specific solutions were Tracks inevitable, but still, many bugs solved already that would otherwise have appeared in 2007! Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
6 Data taking n Beams used: 1 -180 Ge. V pion/electrons/photons A complete vertical slice of ATLAS is available and we can exploit the full power of the ATLAS detector for detailed particle id and measurement Situation where many detector communities interact, and have to work together towards a common physics aim Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
7 Event display Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
8 Event display Welke bundel Welke software Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
9 Event display Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
10 Some events are more difficult Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
11 SCT residuals after alignment Before alignment n After alignment, the average residual over all the planes is found to be 25 micron (again close to expected). Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
12 TRT calibration Without calibration n With calibration The TRT hits are represented by vertical lines. The top and bottom of the line are the edges of the driftcircle, so that’s where the track should be. Without calibration, Athena was assuming t 0 = 0 for all the straws, while in reality it is more like 25 ns. As a result, the residuals are about 1 mm. Calibration clearly improves things. If the straw is not found in the map of calibrated straws then the average of all the straws is applied (still much better than the Athena default) Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
13 Curved tracks with helix fitter Run 2102130, 9 Ge. V pions, B=1. 4 T Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
14 Curved tracks with helix fitter Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
15 Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
16 Ganged pixels n The particle hits almost the same spot in each of the modules it hits. So often you get ganged pixels in three layers at the same time. n Probably need detailed alignment to solve this reliably, these hits are ignored for now in the SL fitter. Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
17 Momentum reco with x. Kalman Run 2102130, 9 Ge. V pions, B = 1. 4 T Alleen pixel en sct Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
18 Inner detectgor: conclusions n n Wat hebben we bereikt tot nog toe? Wat moeten we nog doen? ¡ ¡ ¡ Aligniering Calibratie Conditions: noise levels + dead channels efficiency Algemene performance Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
19 Calorimetry: LAr-Tile correlation e p m Quasi-online monitoring inside Athena Tile and LAr not calibrated Muons removed with Muon-tag Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
20 calorimeter Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
21 Muons: Sagitta measurements Sagitta definition at the TB: • Create r-z and phi track. Outer Middle Sagitta Inner station segments, perform pattern recognition • Segments fit on each muon station • Combination of inner and outer, compare with middle station • Misalignment affects mean value and width • Multiple scattering affects width Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
22 Comparisons with G 4 simulation Sagitta width G 4 Data Athena 8. 7. 0 - Difference G 4 -Data 1/p: material problem (effect corresponding to ~ 3 cm aluminum equivalent). - Detailed check of materials associated to Geo. Model volumes-problem was found in RPC Athena 8. 8. 0 After fixing the problem RPC internal structure DED Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
23 ID-Muon Track Correlation ID ID Muons ID Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
24 Conclusions n n n Was het een leuke/leerzame testbeam periode? Testbeam volgend jaar? Zijn we nu helemaal voorbereid op het nemen van data bij Atlas? Thijs Cornelissen - NIKHEF Jamboree, Dec 2004
- Slides: 24