Performance of CMS hadron calorimeter timing and synchronization using test beam, cosmic ray, and LHC beam data

Performance of CMS hadron calorimeter timing and synchronization using test beam, cosmic ray, and LHC beam data
of 33

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  CMS PAPER CFT-09-018 CMS Paper 2010/01/12 Performance of CMS Hadron Calorimeter Timing andSynchronization using Test Beam, Cosmic Ray, and LHCBeam Data The CMS Collaboration ∗ Abstract This paper discusses the design and performance of the time measurement techniqueand of the synchronization systems of the CMS hadron calorimeter. Time measure-ment performance results are presented from test beam data taken in the years 2004and 2006. For hadronic showers of energy greater than 100 GeV, the timing resolutionis measured to be about 1.2 ns. Time synchronization and out-of-time background re- jection results are presented from the Cosmic Run At Four Tesla and LHC beam runstaken in the Autumn of 2008. The inter-channel synchronization is measured to bewithin ± 2 ns. ∗ See Appendix A for the list of collaboration members   a  r   X   i  v  :   0   9   1   1 .   4   8   7   7  v   2   [  p   h  y  s   i  c  s .   i  n  s  -   d  e   t   ]   1   2   J  a  n   2   0   1   0  1 1 Introduction The primary goal of the Compact Muon Solenoid (CMS) experiment [1] is to explore parti-cle physics at the TeV energy scale exploiting the proton-proton collisions delivered by theLarge Hadron Collider (LHC) [2]. The central feature of the CMS apparatus is a supercon-ducting solenoid, of 6 m internal diameter, providing a field of 3.8 T. Within the field volumeare the silicon pixel and strip tracker, the crystal electromagnetic calorimeter (ECAL) and the brass/scintillator hadron calorimeter (HCAL). Muons are measured in gas-ionization detectorsembedded in the steel return yoke. In addition to the barrel and endcap detectors, CMS hasextensive forward calorimetry.The primary purpose of the HCAL is the measurement of hadronic energy from collisions inCMS. In addition to the energy measurement, the HCAL is also able to perform a precise timemeasurement for each energy deposit. Precise time measurements are valuable for excludingcalorimeter noise and energy deposits from beam halo and cosmic ray muons. Time informa-tion can also be valuable for identifying new physics signals such as long-lived particle decaysand slow high-mass charged particles [3].This paper is organized as follows. Section 2 reviews the pertinent details of the HCAL con-struction and their fundamental impact on timing resolution. Section 3 discusses the methodused to extract a time value from the digitized HCAL signal. In Section 4, the validation of the method is presented based on measurements in test beam and initial beam operations of the LHC in September 2008. Section 5 details the performance of timing filters in the suppres-sion of non-collision-based backgrounds and the effect these timing filters have on simulatedphysics events. 2 CMS Hadron Calorimeter Description A detailed description of the HCAL can be found elsewhere [1]. Briefly, the HCAL consistsof a set of sampling calorimeters. The barrel [4] and endcap [5] calorimeters utilize alternat-ing layers of brass as absorber and plastic scintillator as active material. The scintillation lightis converted by wavelength-shifting fibers embedded in the scintillator and channeled to hy- brid photodiode detectors via clear fibers. The outer calorimeter [6] utilizes the CMS magnetcoil/cryostat and the steel of the magnet return yoke as its absorber, and uses the same activematerial and readout system as the barrel and endcap calorimeters. The forward calorimeter is based on Cherenkov light production in quartz fibers and is not discussed in this paper, due toits different signal time structure.The HCAL is segmented into individual calorimeter cells along three coordinates,  η ,  φ , anddepth. The  φ  coordinate is the azimuthal angle and  η  is the pseudorapidity. The depth is aninteger coordinate that enumerates the segmentation longitudinally, along the direction fromthe center of the nominal interaction region. The layout of the barrel, endcap and outer calori-meter cells is illustrated in Fig. 1. Most cells include several scintillator layers; for example, inmost of the barrel all 17 scintillator layers are combined into a single depth segment.Calorimetric measurements are acquired using the HCAL readout electronics, shown schemat-ically in Fig. 2. Each electronics channel collects and processes the signal of a single cell. Onecalorimetric measurement is acquired with each LHC clock tick (25 ns) from each cell in theHCAL. This defines a “time sample”. The HCAL front-end electronics does not sample thesignal instantaneously; rather, the electric current collected from the photodetectors is inte-grated over each clock period and then sampled. As a consequence, the sample clock is most  2  2 CMS Hadron Calorimeter Description Figure 1: A quarter slice of the CMS HCAL detectors. The right end of the beam line is theinteraction point. FEE denotes the location of the Front End Electronics for the barrel and theendcap. Inthediagram,thenumbersonthetopandleftrefertosegmentsin η ,andthenumbersontherightandthebottomrefertoscintillatorlayers. Colors/shadesindicatethecombinationsof layers that form the different depth segments, which are numbered sequentially startingat 1, moving outward from the interaction point. The outer calorimeter is assigned depth 4.Segmentation along  φ  is not shown. Fiberto OpticalDigitalDataConfigurationLHCClockPhotodiodeHybridQIEADC Delay HCAL cellScintillator LayersData q= Idt Figure 2: A schematic view of the HCAL front-end readout electronics. The readout for oneHCAL cell/channel is shown. Key features are the optical summing of layers, charge integra-tionfollowedbysamplinganddigitization, andper-channelprogrammabledelaysettings. The”QIE” [7] is a custom chip that contains the charge-integrating electronics with an analog-to-digital converter (ADC). The Configuration Data input defines the sampling delay settings.  2.1 Sources of timing variation and uncertainty  3 commonly termed the “integration clock”.The integration clock can be delayed with respect to the LHC clock on a per-channel basis byprogrammable settings, referred to as sampling delay settings, that have a resolution of 1 ns.The purpose of these settings is to compensate for channel-dependent timing variations at thehardware level, and to permit the energy estimate from each LHC crossing to be reconstructedconsistently for use in the Level-1 trigger for all pulse amplitudes and for all channels. Theyalso provide an initial coarse timing calibration. 2.1 Sources of timing variation and uncertainty There are two dominant sources of channel-dependent timing variation: the different time-of-flight of particles from the interaction region to each HCAL cell and the different signalpropagation times through optical fibers of different lengths. Within the barrel, the first effecttends to skew reconstructed times later for higher  η . The second effect, because of the locationof the front-end electronics, tends to skew reconstructed times earlier for higher  η , and this isthe effect that dominates. The combination of these two effects induces a timing dependenceon  η , with a spread in each half-barrel of about 15 ns. Smaller variations as a function of   φ are induced by clock distribution differences and other effects. By applying proper samplingdelaysettings,thisspreadcanbereducedsubstantially. Inaddition,thereconstructionsoftwareutilizes a set of per-channel calibration constants to synchronize the mean timing of energymeasurements to less than 1 ns.Fiber lengths also differ within each calorimeter cell along the radial coordinate, but in this casethere are no means available to compensate for these differences. Since the signals from all thescintillator layers comprising one cell are optically summed, and the optical path lengths arenot equalized across the layers, the resulting signal is smeared in time. For hadrons, whichexhibit large shower-to-shower fluctuations in longitudinal development, the optical summingof layers imposes a limit on the timing resolution, estimated to be approximately 1 ns. For sig-nals with uniform energy deposit in each layer (such as those arising from the beam-collimatorinteractions described later), the resolution is not limited by shower-development fluctuationsand can be significantly smaller than 1 ns. 3 HCAL Time Reconstruction When the Level-1 trigger system [8] identifies an event of potential physics interest, a set of 10 consecutive time samples per channel containing the triggered bunch-crossing is collectedand sent to the high-level trigger software. The capability of the HCAL system to reconstructthe arrival time of the signal more precisely than the sample clock period derives from thespread of the HCAL pulse shape over three to four time-integrated samples (Fig. 3). The timereconstruction software calculates a first order time estimate from a center-of-gravity techniqueusing the three samples centered on the peak,Weighted peak bin  =  (  p − 1 )  A  p − 1  +  pA  p  + (  p  + 1 )  A  p + 1  A  p − 1  +  A  p  +  A  p + 1  × C  , (1)where  A i  represents the amplitude of an arbitrary time sample  i , and  p  is the value of   i  suchthat  A  p  is maximum over the set of samples. In the case of multiple samples with the sameamplitude, the earliest one is picked. The constant  C  is an amplitude-independent normaliza-tion constant that rescales the first order estimate to a range from zero to one. The weighted
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks