Performance of the CMS Hadron Calorimeter with Cosmic Ray Muons and LHC Beam Data

TL;DR

The paper documents a comprehensive commissioning program for the CMS Hadron Calorimeter (HCAL) using test beams, cosmic-ray muons, and LHC beam data. It develops and validates a calibration chain across all HCAL subdetectors (HB, HE, HO, HF), examines magnetic-field induced response changes, and optimizes hardware and electronics behavior via CRAFT data. Key outcomes include ~5% intercalibration precision for the majority of HB channels and ≤12% across other subdetectors, along with observed magnetic-field signal increases and a stable, low-noise environment suitable for early LHC data-taking. The results underpin HCAL performance and guide ongoing calibration strategies for accurate jet and missing energy measurements at the LHC.

Abstract

The CMS Hadron Calorimeter in the barrel, endcap and forward regions is fully commissioned. Cosmic ray data were taken with and without magnetic field at the surface hall and after installation in the experimental hall, hundred meters underground. Various measurements were also performed during the few days of beam in the LHC in September 2008. Calibration parameters were extracted, and the energy response of the HCAL determined from test beam data has been checked.

Figures (13)

• Figure 1: The CMS HCAL detector (quarter slice). "FEE" indicates the locations of the Front End Electronics for HB and HE. The signals of the tower segments with the same color are added optically, to provide the HCAL "longitudinal" segmentation. HB, HE and HF are built of 36 identical azimuthal wedges ($\Delta \phi$ = 20 degrees).
• Figure 2: Cosmic ray muon energy deposition measured in HB (left) and HE (right) with $B = 3.8$ T for muon momenta above 7 GeV/$c$. HB and HE data samples consist of 450 and 27 thousand events, respectively.
• Figure 3: Ratio of the scintillator signals in a single layer with magnetic field on and off, measured using a radioactive source during MTCC. The left figure is for polystyrene (PS) in HB and the right figure is PVT (filled squares) and PS (open squares) in HE. The horizontal lines represent the averages of each data set. The difference for PS between HB and HE is due to the curling of the knock-on electrons and the magnetic field direction.
• Figure 4: Left: Magnetic field strength at the location of the HO scintillators, as calculated with the TOSCA program bib:tosca. The vertical lines indicate the ring borders. Ring 0 has two scintillator layers, at radii 3.82 m and 4.07 m, respectively. Right: Pedestal and cosmic ray muon signal in Ring 0 of HO.
• Figure 5: Total noise rate for around 4000 trigger towers as a function of the energy threshold, during CRAFT with the magnet at full field. A minimum energy of 20 GeV was required in at least one HB or HE readout box (RBX). About 280 HPDs, corresponding to 5040 channels of HB and HE, are included in the analysis; HPDs that were replaced in the beginning of 2009 are excluded.