IntroductionCMS PbWO4 ECAL Resolution δLO/LO versus δLT/LT @ 100 rad/hLaser Source SpecificationsContinuous Monitoring in situLaser Light Monitoring SystemLaser System Commission at CERNNd:YLF pumped Ti:Sapphire LaserLaser Light Distribution SystemLaser DAQ and Distribution SystemLasers Performance MonitoringMonitoring System Stability Beam Irradiation Tests Laser Monitoring is Effective Summary1October 29, 2007N09-4, Kejun Zhu, CaltechThe CMS ECAL Laser Monitoring SystemKejun ZhuCalifornia Institute of TechnologyOn Behalf of the CMS ECAL Group2October 29, 2007N09-4, Kejun Zhu, CaltechIntroduction¾At the LHC design luminosity, the CMS ECAL is exposed in a harsh radiation environment (dose rates of 15 rad/hour for the barrel and 500 rad/h for the endcaps at 1034cm-2s-1). The lead tungstate (PbWO4) crystals suffer from a dose-rate dependent radiation damage. ¾Radiation causes a degradation in crystal transparency because of color center formation. The crystals recover in absence of radiation with two time constants ranging from days and weeks. Changes in crystal transparency, and therefore calorimeter response, must be corrected for to maintain the ECAL resolution.¾A laser based light monitoring system is designed to measure the transparency variations of each crystal continuously during LHC running with 0.2% precision.3October 29, 2007N09-4, Kejun Zhu, CaltechCMS PbWO4ECAL ResolutionBeam TestDesigned Resolution 75,848 PbWO4crystals To maintain 0.5% constant term in energy resolution, the monitoring precision is required to be 0.2%.4October 29, 2007N09-4, Kejun Zhu, CaltechδLO/LO versus δLT/LT @ 100 rad/hStrong correlation observed between variations of the light output and the longitudinal transmittance for full size PbWO4crystals in multi cycles of irradiation and recovery5October 29, 2007N09-4, Kejun Zhu, CaltechÆ 2 wavelengths per laserÆ Pulse FWHM: < 40 ns to match ECAL readoutÆ Pulse jitter: < 3 ns for synchronization with LHCÆ Pulse rate: ~100 Hz, scan of full ECAL in 20minÆ Pulse intensity instability: < 10%Æ Pulse energy: 1 mJ/pulse at 440 nm, equivalent to 1.3 TeV in dynamic rangeLaser Source Specifications6October 29, 2007N09-4, Kejun Zhu, CaltechContinuous Monitoring in situUsing 1% beam gaps (100 Hz) in the LHC beam structureScan entire ECAL in 20 minutes7October 29, 2007N09-4, Kejun Zhu, CaltechLaser Light Monitoring SystemTwo lasers to guarantee 100% availability of 440 nm8October 29, 2007N09-4, Kejun Zhu, CaltechLaser System Commission at CERN1stlaser: 8/20012nd& 3rdlasers: 8/2003Two lasers at P5: 3/2007Each laser system contains an Nd:YLF pump laser and a tunable Ti:Sapphire laser with dual wavelengths9October 29, 2007N09-4, Kejun Zhu, CaltechNd:YLF pumped Ti:Sapphire LaserTunable Ti:S Nd:YLF PumpLasers operate at two wavelengths by using interference filters in pathBlue/green: 440/495 nmIR/Red: 796/709 nm10October 29, 2007N09-4, Kejun Zhu, CaltechLaser Light Distribution SystemFibers at front for barrel crystalFibers at back for endcap crystalAn optical fiber based two-level light distribution system designed and constructed by the Saclay group Integrating sphere based level-1 distribution for good uniformity11October 29, 2007N09-4, Kejun Zhu, CaltechLaser DAQ and Distribution System12October 29, 2007N09-4, Kejun Zhu, CaltechLasers Performance MonitoringClass < 1,00004/2005An Acqiris DP210 card of 2 GS/s was installed in 2004 to provide pulse energy, FWHM and timing information for each laser pulse. Short (30 min)/long (25 h) term stabilities: < 2%/3%Lamp aging: 0.5% daily, leading to long term degradation.13October 29, 2007N09-4, Kejun Zhu, CaltechPerformance published in IEEE Trans. Nucl. Sci. vol. 52 pp. 1123-1130 (2005): 25 h stability of pulse energy & width: 3% and a long term degradation of laser pulse timing. Software Feedback to Reduce Jitters¨ Laser pulse intensity, width and timing are correlated to the YLF pumping current.¨ Better pulse stability could be achieved by trimming the YLF laser pumping current.14October 29, 2007N09-4, Kejun Zhu, CaltechA factor of two better stability in laser pulse energy and width and a jitter of less than 2 ns were observed in laser runs lasting for about three months, when YLF current increased automatically at 0.1 nA steps.Performance with Software Feedback15October 29, 2007N09-4, Kejun Zhu, CaltechMonitoring System StabilityTypically ~0.1 % long term stability in real beam test environment, including the stability of the entire readout chain - temperature, HV, etc.Transparency variation can be measured to 0.1%From test beam : RMS of APD/PN ratio per channel no irradiation, 450 h.Single channel response16October 29, 2007N09-4, Kejun Zhu, CaltechBeam Irradiation Tests17October 29, 2007N09-4, Kejun Zhu, CaltechLaser Monitoring is Effective 120 GeV electrons reconstructed by 3x3 crystal matrix in irradiation test18October 29, 2007N09-4, Kejun Zhu, CaltechSummary A laser monitoring light source was designed and constructed at Caltech for the CMS ECAL, and was installed and commissioned at CERN since 2001.  The 25 h laser performance satisfies the original specifications. With a software feedback, the long term (3 months) performance of the laser system also exceeds the design specifications. The laser monitoring system has been used in the CMS ECAL beam tests since 2001. Irradiation test beam data demonstrated that the laser monitoring is effective in maintaining the intrinsic crystal calorimeter