Laser monitoring system for the cms lead tungstate crystal calorimeter

Скачать 182.98 Kb.
НазваниеLaser monitoring system for the cms lead tungstate crystal calorimeter
Размер182.98 Kb.
  1   2   3   4   5   6   7


Laser monitoring system for the CMS lead tungstate crystal calorimeter

M. Anfrevilleb, D. Bailleuxa, J.P. Bardb, A. Bornheima, C. Bouchandb, E. Bougamontb, M. Boyerb, R. Chipauxb, V. Daponte-Puillc#, M. Dejardinc*, J.L. Faurec, P. Grasc, P. Jarryc, C. Jeanneyb, A. Joudonb , J.P. Pansartc, Y. Penichotb, J. Randerc, J. Rolquinb, J.M. Reymondb, J. Tartasb, P. Venaultb, P. Verrecchiac , L. Zhanga, K. Zhua, R.Y. Zhua

aCalifornia Institute of Technology, Pasadena, CA 91125, USA

bCEA, DAPNIA/SEDI, CE-Saclay, F-91191 Gif-sur-Yvette CEDEX, France

cCEA, DAPNIA/SPP, CE-Saclay, F-91191 Gif-sur-Yvette CEDEX, France

*Corresponding author. E-mail address:

#Now at LAL, IN2P3-CNRS, Orsay, France.


We report on the multiple wavelength laser monitoring system designed for the CMS lead tungstate crystal calorimeter read-out with avalanche photodiodes (Barrel calorimeters) and vacuum phototriodes (End Cap calorimeters). Results are presented for the test beam performance of the system designed to achieve 0.5% relative inter-calibration of the optical transmittance for lead tungstate scintillation emission over nearly 80 000 channels. The system operates in continuous measurement cycles to follow each crystal’s evolution under irradiation and recovery periods foreseen during operation at the LHC.

Keywords: Laser, Monitoring system, Optical transmittance, Stability, Calorimeter, Scintillating crystal

  1. Introduction

Lead tungstate (PbWO4) crystals were chosen for 75 848 channel electromagnetic calorimeter (ECAL) currently under construction for the Compact Muon Solenoid (CMS) experiment at the CERN LHC [1]. This choice was based upon the crystal’s high density and the intrinsic radiation hardness of the scintillation light mechanism. The scintillation signal Si for a single channel (i) at a given emission wavelength λ in such a crystal calorimeter can be factorized approximately as follows:

Si(E, λ) = [N(E) · LYi(λ)] · Tri (λ) · [A i · QE i(λ) ·M i(λ)] ,

where the first term includes the shower deposition N(E) and scintillation light yield factor LY, the second term is the optical transmission Tr at the given wavelength, and the third term regroups the geometrical acceptance A, the quantum efficiency QE, and gain M of the photodetector. The first term is unaffected by irradiation at LHC, and the third term can be controlled by designing radiation hard photodetectors and by measuring the electronic gain. The optical transmission, however, is a critical issue for these crystals, since light transmission at the scintillation wavelengths is affected by the production of color centers under electromagnetic irradiation [2][3]. Furthermore, the annealing of these color centers at room temperature leads to a transmission recovery. Light transmission at any moment of time is the result of equilibrium between the rates of color center production and annealing. Crystals produced for ECAL are the result of a long R&D process to optimize doping and production stoechiometry to improve radiation hardness, thereby reducing the scale of the variations in light transmission to less than 6 % under LHC conditions (γ irradiation dose rates of 0.15 Gy/h at luminosities of 1034 cm-2 s-1).

A critical issue in the performance of such a large system is the cell-to-cell uniformity. The energy resolution in the reconstruction of the Higgs two-photon decay is seriously degraded if the light transmission term is uncertain at the few % level. Final calibration of the ECAL will be achieved using physics events collected over days or weeks, depending upon the calibration process, but the energies need to be corrected for the short-term transmission variations. The goal of the system described in this paper is to achieve ≤ 0.5% relative inter-calibration and long term stability (≤ 0.15% drift/month) in the transmission term over the nearly 80 000 channels of the full ECAL. The independent issues of precision and quality control have been addressed at each step of the systems conception and production.

Two options have been used elsewhere for light transmission measurements in large calorimeter systems: radioactive sources or external light injection. Radioactive sources are interesting in that they monitor the product of the light yield and transmission terms. Unfortunately, the relatively high noise level ~ 40 MeV of the ECAL Barrel avalanche photodiode (APD) and its associated electronics did not allow such a solution. Previously designed external light injection systems have used either stable light emitting diodes LED, which were too weak for the CMS application far within a 4 Tesla solenoid, or laser injection systems, which had sufficient power but had not achieved a precision ≤ 1.5%. We have followed this last approach, placing both a light-mixing step and a stable reference monitor sufficiently close to the final light injection at the crystal level to achieve a substantial increase in precision. The use of a laser monitoring system allows frequent (every ~20 min) in situ measurements of each crystal’s light transmission, thereby following the damage and recovery during the physics data taking periods. The transmission results can then be used to correct raw energy deposits measured in the calorimeter in the same time intervals. Test beam studies at CERN have demonstrated the feasibility of maintaining energy resolution with light monitoring [4].

In this paper, we report on the design, installation at the detector level, and performance of the CMS ECAL laser monitoring system. The system has been used during test beam studies of ECAL Barrel supermodules (assembly of 1700 crystals); results relevant to the performance of the laser monitoring system are also presented here.

  1. Laser monitoring system

    1. Overview

The major components of the laser monitoring system presented in this paper are shown schematically in Fig.1. Laser light is produced at a source installed in the CMS underground service cavern USC55 using tunable Ti:Sapphire lasers operating at either of two wavelengths: 440 nm, the principal wavelength, at the Y doped PbWO4 scintillation emission peak (where 495 nm near the undoped PbWO4 emission peak is kept as a cross-check) or 796 nm (where 706 nm is also available), only weakly sensitive to color center production, used as a cross-check of gain variations. Pulse energy at the principle wavelength is 1mJ/pulse, at the source, corresponding to the scintillation light of ~1.3 TeV energy deposit in a crystal, where the total attenuation of the light distribution system is measured to be 69 dB (laser to crystal front face). An attenuator allows 1% steps down to an equivalent energy of ~13 GeV. The laser light pulses are directed to individual crystals via a multi-level optical fiber distribution system: a) a fiber-optic switch at the source directs the pulses to one of 88 calorimeter elements (ECAL Barrel: 72 half supermodules, ECAL End Cap: 16 quarter Dee’s), b) a primary optical fiber distribution system transports the pulses 95 to 130 m to the each calorimeter element mounted in CMS located in the experimental cavern UXC55, finally, a two-level distribution system mounted on the detector sends the pulses to the individual crystals. The basic principle of operation is illustrated for the ECAL Barrel geometry in Fig. 1: laser pulses transported via an optical fiber are injected at a fixed position at the crystal’s front face, the injected light is collected, as similarly done for scintillation light from an electromagnetic shower, using a pair of avalanche photodiodes (APD) glued to the crystal’s rear face, and the pair are read-out in parallel with the front-end electronics chain. This design ensures that the crystal’s optical transmission is measured in the region of interest, although the optical light path is somewhat different from that taken by scintillation photons. The underlying principle is similar for ECAL End Cap, however there the calorimeter design is based on identical 5x5 crystal units [1] which does not permit front face injection; in this case, laser light is injected at a corner of each End Cap crystal’s rear face, and the light is collected (as for scintillation) via a vacuum phototriode (VPT) glued on the crystal’s rear face. Since the optical transmission depends upon light path, Tr(λ) differs between Barrel and End Caps. In order not to interfere with ECAL performance during physics collisions at LHC, the laser pulses are injected during 3.17 μs gaps foreseen every 88.924 μs in the LHC beam structure [5], as shown in Fig. 2. The laser monitoring system independently measures the injected light for each pulse distributed to a group of typically 200 crystals using pairs of radiation hard PN photodiodes read-out via dedicated front-end electronics. The essential quantity used to make crystal optical transmission corrections is the ratio of the appropriate APD (or VPT for the End Caps) response normalized by the associated group’s PN response. Altogether, 75 688 crystals, 61 200 for ECAL Barrel and 14 488 for the ECAL End Caps, will be monitored by the system at LHC, cycling continuously over the 88 calorimeter elements. The components of the monitoring system mounted on the calorimeter are designed to be radiation hard; full LHC luminosity running over 10 years, depending upon angle, will expose them to γ doses of typically 3 kGy, and neutron fluences of 2x1013 n/cm2 [6]. Irradiation test results are discussed in section 3.

Figure 1. Schematic of laser monitoring system. Laser pulses at either of two wavelengths are distributed to calorimeter elements via an 1:88 optical switch and 90-130 m long primary fibers. The pulses are distributed on the detector via a secondary (L2) fan-out which subsequently serves 5 (ECAL Barrel) tertiary (L1) fan-outs, injecting the light into groups of typically 200 crystals. Signals from pairs of reference PN photodiodes which monitor light from the fan-outs are amplified in front-end modules (FEM), and are in turn digitized in the MEM before being sent to the CMS ECAL DAQ. Laser pulses in each PbWO4 crystal are detected and amplified as for scintillation light from electromagnetic showers via twin avalanche photodiodes (APD) and are treated by the CMS ECAL electronic chain’s Very Front End card (VFE).

Figure 2. Laser pulse timing diagram showing the LHC beam structure and the laser pulse within the 3μs abort gap (without physics collisions) foreseen every 88.924 μs cycle to allow for the rise time of the beam dump’s fast kickers. A latency of 5.6 μs is required between the Test Enable signal and the arrival of the laser pulse at the calorimeter. Sufficient time is allowed for last physics event buffer readout before the laser trigger is sent.

Figure 3. Schematic diagram of monitoring light source installed in the laser barrack in the CMS underground service cavern.

    1. Multiple wavelength monitoring light source and high level distribution system

Since the monitoring light source has been presented elsewhere [7] only the design requirements and the essential characteristics are reviewed here. The laser source specifications and its environmental requirements are listed below:

  • Operation duty cycle: 100% during LHC data taking periods (expected to be about 5000 hours/year).

  • Two operating wavelengths: 440 nm (near PbWO4 emission peak) & 796 nm (for electronics cross-checks).

  • Spectral contamination: < 10-3, which was confirmed with a high resolution monochromator, Oriel MS257.

  • Pulse energy (Epulse): 1mJ at the source for a dynamic range up to 1.3 TeV in a crystal with 69 dB attenuation in the distribution system.

  • Pulse width (Γpulse): < 40ns FWHM to match the ECAL readout. See sections 2.4 and 5.2 for discussion.

  • Pulse jitter: < 4 ns (24 hours), < 2 ns (30 min).

  • Pulse rate: ~ 80 Hz, which is allowed by ECAL DAQ.

  • Pulse to pulse instability: < 10% of Epulse.

  • Clean room class: better than 10 000 to protect laser optics.

  • Temperature stabilization: ±0.5° C and humidity <60%.

The monitoring light source consists of three laser systems (two “active” and one “spare”) each equipped with diagnostics, shown schematically in Fig. 3. Each laser system mounted on its optical bench, see Fig. 4, consists of two lasers: 1) a Quantronix [15] model 527DQ-S Q switched Nd:YLF pump laser, which delivers pulses up to 20 mJ at 527 nm, and 2) a Quantronix [15] custom made Proteus UV(SHG) tunable Ti:Sapphire laser, supplying up to 1 mJ pulses at one of two available wavelengths, and at rates up to 100 Hz. Laser characteristics are summarized in Table 1. Typical pulse shapes are shown in Fig. 5.

Table 1. Monitoring Light Source Laser Characteristics.




Ti:Sapphire 1

Ti:Sapphire 2

Wavelength λ (nm)






Pulse energy (mJ)






Pulse width (ns)






The output of either one of two active systems operating at the principal wavelengths, 440 nm and 796 nm, or the spare at 440 nm is selected with a DiCon [16] 3x1 fiber-optic switch. The output is intensity is adjusted by two attenuators in series: a 1% step linear attenuator and a logarithmic attenuator. Both are controlled by computer, allowing pulse intensity scans to measure linearity. Laser pulse energy and FWHM are monitored in the laser barrack using PIN photodiodes and Digital Sampling Oscilloscope sampling at 1 Hz. The attenuated pulses are sent to the selected calorimeter element’s primary distribution fiber via a computer controlled DiCon [16] 1x88 fiber-optical switch.

Figure 4. Monitoring system multiple wavelength light source (shown installed at CERN H4 test beam area). Light pulses at 440 (or 495) nm are produced with a tunable Ti:Sapphire laser(front) using a Q-witched 527 nm Nd:YLF pump laser (back). A second laser system provides light pulses at 796 (or 709) nm.

Figure 5. Measured pulse shapes for monitoring light source Nd:YLF pump laser at527 nm and Ti:Sapphire laser at 440 nm.

  1   2   3   4   5   6   7


Laser monitoring system for the cms lead tungstate crystal calorimeter iconThe Place and Role of Official Statistics within the National System of Health Monitoring

Laser monitoring system for the cms lead tungstate crystal calorimeter iconTitle: sens-u: Remote human in loop health-monitoring system at home

Laser monitoring system for the cms lead tungstate crystal calorimeter iconLaser-Ablation Rates Measured Using x-ray Laser Transmission

Laser monitoring system for the cms lead tungstate crystal calorimeter iconStatistical Support and Web Development for a Web-based Master Sample Management System for Integrating Aquatic Ecosystem Status and Trend Monitoring

Laser monitoring system for the cms lead tungstate crystal calorimeter iconПрименение cms для организации самообразовательной деятельности студентов
Гибкое управление самообразованием учащихся и педагогов в области интернет – технологий и обеспечение учебного процесса необходимыми...
Laser monitoring system for the cms lead tungstate crystal calorimeter iconLead metal cas no: 7439-92-1 einecs no: [click here to insert einecs no.] Lead oxide

Laser monitoring system for the cms lead tungstate crystal calorimeter iconМетодика разработки инструментария для администрирования веб-сайтов Бондаренко В. Ю. Руководитель Алексеев Н. А., к т. н
Эта разработка была внедрена и протестирована на веб-проектах сравнительного анализа систем управления контентом Findcms org ua (cms,...
Laser monitoring system for the cms lead tungstate crystal calorimeter iconElectronic Signatures and Infrastructures (esi); cms advanced Electronic Signatures (CAdES)
Электронные подписи и инфраструктуры (esi); усовершенствованные электронные подписи cms (CAdES)
Laser monitoring system for the cms lead tungstate crystal calorimeter iconCrystal Structure of Natural Zeolites

Laser monitoring system for the cms lead tungstate crystal calorimeter iconSystem informacyjny system informatyczny. Co poza nazwą RÓŻni te dwa obiekty?

Разместите кнопку на своём сайте:

База данных защищена авторским правом © 2014
обратиться к администрации
Главная страница