ATLAS Calibration and Monitoring Systems June 19 th -23 rd, 217 Arely Cortes-Gonzalez (CERN) On behalf of the ATLAS Collaboration
ATLAS Detector Trigger Hardware based L1 ~1kHz Software based HLT ~1kHz Muons Spectrometer MS includes precision tracking chambers ( η <2.7) and fast detectors for triggering ( η <2.4). Calorimeters LAr and scintillator calorimeters provide EM and hadronic energy measurements up to η <4.9 Tracking Inner tracking detectors cover up to η =2.5. 2
The ATLAS Tile calorimeter (TileCal) is the central section of the hadronic calorimeter. o Sampling calorimeter using iron plates as absorber and scintillating tiles as active medium. o 3 sections: Long Barrel (2 readout regions: LBA, LBC) and two Extended Barrels (EBA, EBC). o Long Barrel: < ỷ < 1. o Extended Barrel:.8 < ỷ < 1.7 o The readout is segmented into ~5 cells (longitudinally and transversally), each read by two PMTs. TileCal provides important information for reconstruction of hadrons, jets, hadronic decays of tau leptons and missing transverse energy. Hadron energy resolution for jets:. 3
Each partition is divided into 64 symmetric φ slices (modules), with 45 instrumented channels in LB modules and 32 channels in EB modules. η and radial structure of TileCal cells o Radially: 3 longitudinal layers (total thickness of ~7.4λ). o A-cells closest to beam axis, followed by BC- and D-cells. o E-cells installed in the gap/crack region (1. < ỷ < 1.6). o ỷ φ granularity: ỷₒ φ =.1ₒ.1 Photomultiplier Optical readout of signals in TileCal: o Light produced in scintillating tiles. o Readout via WLS fibres connected to both edges of the scintillating tiles. o Converted into electric currents by the PMTs. o Their signal is shaped and amplified with two gains (LG, HG). Dynamic range: ~1 MeV to ~8 GeV. Wave-length shifting fiber Scintillator Steel Source tubes 4
Calibration Systems Particles Calorimeter Tiles Photomultiplier Tubes Integrator Readout (Cs & Particles) Digital Readout (Laser & Particles) 137 Cs source Laser light Charge injection (CIS) The reconstructed energy is derived from the raw response: Cell response is not constant in time due to the PMT gain variation and scintillator degradation due to the exposure to beam. Systems used for calibration in TileCal o Charge Injection System (CIS): o Calibrates the response of ADCs (electronics): C ADCàpC. o Cesium system: o Calibrates optical components and PMT gains: C Cesium. o Laser System: o Calibrates variations due to electronics and PMTs: C Laser. o Minimum Bias System (MB): o Monitors beam conditions, TileCal optic components and PMT gains. o C pcàgev : measured during dedicated test beam campaigns. 5
Integrator Readout Particles Calorimeter Tiles Photomultiplier Tubes Integrator Readout (Cs & Particles) Used by Cesium and MB Digital Readout (Laser & Particles) 137 Cs source Laser light Charge injection (CIS) The integrator system integrates the PMT signal over a large time, providing a slow control readout. o The integrator is printed on a circuit board plugged to the so-called 3-in-1 card. o A 12-bit ADC card digitises the integrator output which ranges up to 5 V. o The integrator gain can be varied by selecting one among six resistors that also define the integration time (between 1-2 ms). o The integrator gains are configured depending on the instantaneous luminosity for the MB signals. o The average gain stability has been better than.1% during the first run of the LHC. 6
Charge Injection System o Calibrates the response of ADCs (electronics): digital gains and linearities. o Injecting a known charge and measuring pc/adc. o Spanning the full ADC range (-8 pc; 1 pc~1 GeV). o Monitors both LG and HG for all channels. o Allows to simulate a physics signal from the PMT. o Also used to calibrate analog L1 calorimeter trigger. o Extract the conversion factors from ADC counts to pc (C ADCàpC ). o Calibrations taken from daily to weekly. o CIS systematic uncertainty ~.7%. o The stability of the calibration factors is at the level of.3-.4% (for both gains). o Less than 1% of TileCal channels exhibit large fluctuations. 7
Cesium System o The Cesium system is based on three moveable radioactive sources (t đ ~3 years) using a hydraulic control (through a system of steel tubes). o The 137 Cs ữ-sources move inside the calorimeter, emitting.662 MeV photons to illuminate the scintillators. o It uses the integrator readout system during source movement. o Calibration of the complete optical chain (scintillator tiles, fibres, PMTs) and monitoring of the detector response over time: C Cesium. o Between Run I and Run II: improvement of stability and safety of the operation (new water storage system, lower pressure, precise water level metering). 8
Cesium System Variation of TileCal channels response measured in Cesium calibration runs. Scans in Feb-215 used to recalibrate and equalize 2/13 7/14 LHC Long Shutdown o Cell response is not constant in time due to the PMT gain variation and scintillator degradation due to the exposure to beam. o Precision of a single tile response measurement in the order of.5%. o Small deviations from zero observed due to magnetic field (up to.7% differences in scan w/ or w/o field). o Larger deviations in time observed for cells closer to the beam line. 9
Laser System o The gain stability of each PMT is measured using the Laser system. o PMT gain drifts affects the detector response and thus should be measured regularly. o The system sends a controlled amount of light into each PMT (532 nm green light). o Deviations of any channel response wrt nominal is translated into a calibration constant: C Laser. o Laser pulses also sent during collision runs (empty bunches), used to calibrate the calorimeter timing. o During the LHC Long Shutdown between Run I and Run II a new Laser II system was developed to correct shortcomings in electronics and light monitoring from the original system. o Upgraded optics box and control electronics for Run II. o Improved laser light estimation (more photodiodes). o Improved resolution: precision on gain variation measurement <.5%. 1
ATLAS Preliminary Laser System..1.2.3.4.5.6.7.8.9 1. 1.1 eta 1.2 D D1 D2 D3-1.15 D4-2.74 1.3 C1-1.65 C2-1.7 C3-1.76 C4-1.76 C5-1.93 C6-1.74 C7-2.2 C8-2.75 C1 D5-2.74-3.4 D6 -.68 1.4 B1-1.65 B2-1.7 B3-1.76 B4-1.76 B5-1.93 B6-1.74 B7-2.2 B8-2.75 B9-2.41 E1-7.31 B11-3.56 B12-3.32 B13-2.48 B14-1.53 B15 -.88 1.6 A1-3.63 A2 A3 A4 A5 A6 A7-3.63-3.82-3.69-3.84-4.21-4.8 A8-4.47 A9-4.55 A1-4.7 E2-1.35 A12A13-4.7-6.14 A14-5.15 A15-3.81 A16-2.53 PMT gain drift <-5.% -2.5% % 2.5% >5.% run 311556, 216-1-27 o Precision on gain variation measurement <.5%. o Allows the monitoring of gain and stability of the PMTs. o Cross check problems (e.g. unstable HV). o Since 216, updates to calibration constants are done as often as weekly, to track changes in PMT responses. o As for Cesium measurements: larger deviations in time observed for cells closer to the beam line. In this case E-cells are also monitored with the Laser system. E3-9.84 E4-14. 11
Minimum Bias System Current [na] 4 1 3 1 2 1 1 1 1 1 ATLAS Preliminary Data 216 s = 13 TeV, L =.2 1 A cells BC cells D cells E cells 33 cm -2 s -1 o High energy proton-proton collisions are dominated by soft parton interactions (MB events). o The integrator readout measures integrated PMT signals over a large time ( 1 ms). o As the Cesium system, the MB system monitors the full optical chain. As of 216 it was used to calibrate in between scans (part of C Cesium ). 1 2 1.5 1.5.5 1 1.5 η cell o Measured currents are linearly dependent on the instantaneous luminosity. o The system can then be used to monitor the instantaneous luminosity. o Or provide an independent measurement given an initial calibration (luminosity coefficient computed from a single run). EBC,mod 22,ch 17 current [na] Current/f(x) 44 42 4 38 36 34 32 3 28 1.1 1.99 Data 215, s=13 TeV Pol1 fit: f(x)= 13.455x+1.32 ATLAS Preliminary 2.2 2.4 2.6 2.8 3 33 Inst. Luminosity [1 cm -2 s -1 ] 12
Combined Calibrations Detector response variation for a single channel in MB is done using a reference cell (typically a cell in the outer D-layer, as a measure of the luminosity). For each run the mean of this distribution is used to extract the response variation for each channel wrt the reference. 215 data taking period o Comparison of response variation between Cesium, Laser and MB measurements. o Cs and MB access PMT gain drift and scintillator ageing. o Laser only monitor PMT gain drifts. o Down-drifts observed during collision periods. o Up-drifts during maintenance periods (recovery of PMTs). o Given the agreement, no ageing (scintillator irradiation observed during 215 alone). A13/D5 Average response [%] -1-2 -3 4/6 215 A13: 1.2 < η < 1.3 D5:.9 < η < 1.1 Laser Minimum Bias Cesium 4/7 215 3/8 215 2/9 215 ATLAS Preliminary 215 Data s = 13 TeV -1 Total Delivered: 4.34 fb 2/1 215 Time [dd/mm and year] 1/11 215 4 35 3 25 2 15 1 5 ] -1 Total Integrated Delivered Luminosity [pb 13
Combined Calibrations 2 4 ATLAS Preliminary 35 216 Data s = 13 TeV 3-1 A13 1.2 < η < 1.3 inner layer D6 1.1 < η < 1.3 outer layer Total Delivered: 39 fb 25 2 Laser 6 15 Minimum Bias 8 1 1 Cesium Scan 29/4 216 29/5 216 5 28/6 216 28/7 216 27/8 216 26/9 216 26/1 Scintillator Irradiation Effect [%] Total Integrated Delivered Luminosity [pb-1] A13/D6 Average response [%] 216 data taking period 2 216 Time [dd/mm and year] 2 4 6 8 1 12 Inner Cells A13: 1.2 < η < 1.3 Gap/Crack Cells E1: 1. < η < 1.1 ATLAS Preliminary E2: 1.1 < η < 1.2 E3: 1.2 < η < 1.4 216 Data s = 13 TeV E4: 1.4 < η < 1.6 Total Delivered: 39 fb-1 12 13 14 15 Integrated Charge [mc] o Differences between Laser and MB measurements can be interpreted as a scintillator ageing due to irradiation. o During 216 this effect was clearly observed for some of the most irradiated cells in the A-layer and E-cells (larger instantaneous and integrated luminosity). o Several ways to charaterise the irradiation effect. o Above: wrt integrated charge, where k is the luminosity coefficient. o Higher irradiation observed for higher collected charges. 14
Detector Status & Data Quality φ TileCal monitoring includes identifying and masking problematic channels correcting for timing jumps, monitoring data corruption or other hardware issues. During maintenance periods (hatched areas) there is an intense campaign to fix all issues (from high to low priority), allowing for a good recovery of the system. Amount of Tile Masked Cells 217-6-3 2 ATLAS Preliminary Evolution of Masked Channels and Cells: 217-6-3 3 [%] 2.5 6 5 4 3 2 1 ATLAS Preliminary Masked Channels (1.18%) Masked Cells (.79%) Maintenance Dec-1 Oct-11 Aug-12 May-13 Mar-14 Jan-15 Oct-15 Aug-16 Jun-17 Redundancy of cell readout system reduces the impact of masked channels. 2 (ỷ,φ) representation of the # of Tile masked cells 1.5 1.5.5 1 1.5 η 2 1.5 1.5 Tile achieved 99.6% data quality efficiency in 212, 1% in 215 and 98.8% in 216, with minor (< 1% of cells) losses by the end of October, 216. 15
Conclusions o TileCal provides important information for reconstruction of hadrons, jets, hadronic decays of tau leptons and missing transverse energy. o Multiple systems are used to calibrate and monitor the response of the TileCal cells. o These calibration systems allowed to achieve great performance of the calorimeter during the LHC Run I and Run II. o Stability of absolute energy scale better than 1%. o Cell response variation in time is tracked with these systems continuously. o This variation is due to PMT gain variation and scintillators degradation due to irradiation. The former recovers during quiet periods without collisions. o These are compensated via calibration constants. o Combining information from the different systems allows to derive the irradiation effect (Laser vs Cesium or Laser vs MB). o This effect is larger at higher integrated luminosity and collected charge. o MB system also allows to monitor instantaneous luminosity for the detector. 16
Many thanks!
Bonus Slides
Calibration Systems 19
Charge Injection System 2
The variation of the response to minimum bias, ATLAS Preliminary 4 cesium and laser for cells in the inner of Tilelayer Calorimeter the Extended Barrel, covering the region " 35 215 Data s = 13 TeV 1.2 < η < 1.3, as a function of the time." 2 1-1 -2-3 -4 EB Cells A13-5 -6 2/3 212 Minimum Bias Integrator Cesium Laser 2/5 212 1/7 212 ATLAS Preliminary -1 Ldt = 22.8 fb, s=8 TeV 31/8 31/1 31/12 212 212 212 Time [dd/mm and year] Total Delivered: 4.34 fb-1 3 Minimum bias data cover the period from the -1 25 beginning of April to the beginning of December 212. The Cesium and Laser results cover the 2 < η < 1.3 to mid-december. The period A13: from1.2 mid-march.9 < time η < 1.1 variation D5: versus for the response of the 3 15 Laser systems is normalised to the first Cesium scan -2 (mid-march, before the start of collisions data 1 Minimum Bias taking). The integrated luminosity is the total 5 delivered during the proton period." Cesium As already observed in 211 the down-drifts of the -3 PMT (seen by 3/8 Laser) coincide with2/1 the 4/6 gains 4/7 2/9 1/11 215 215 215 215 215 215 collision periods while up-drifts are observed during Time [dd/mm and year] machine development periods and at the end of the 1 proton data-taking (beginning of December)." A cells, EB ATLAS Preliminary The drop 8 in the response variation during the data 1.2of η<1.3 Tile tends Calorimeter taking periods to decrease as the exposure 1.3 η<1.4 6 increases. The biggest the PMTs drop is observed -1 s=13 TeV, 4.34 fb 1.4 η<1.5 between 1 and 6 fb-1 and is due to PMT gain down4 1.5 η<1.7 drift. " 2 The Cesium and minimum bias variations are similar, both measurements being sensitive to PMT drift and scintillator irradiation. The difference between 2 minimum bias (or Cesium) and Laser is interpreted as an effect of the scintillators 4 irradiation." The errors bars correspond to the systematic 6 uncertainty summed in quadrature with the module 8 by-module variation and the statistical fluctuation (the module-by-module and statistical fluctuations 1 5 1 15 2 25 3 35 4 45 dominate). Scintillator Irradiation Effect [%] Cell A13 response variation [%] " Total Integrated Delivered Luminosity [pb-1] A13/D5 Average response [%] Combined Calibrations 21 Integrated Charge [mc]