Commissioning and Performance of the CMS Silicon Strip Tracker with Cosmic Ray Muons

TL;DR

The paper documents the commissioning and in-situ performance of the CMS Silicon Strip Tracker using cosmic ray muons in a 3.8 T magnetic field during the CRAFT run. It details the end-to-end commissioning of the SST control and readout systems, including synchronization, gain calibration, pulse-shape tuning, pedestal/noise calibration, and external-trigger alignment, achieving 98% module operation. The SST demonstrates hit efficiencies above 99% and track-reconstruction efficiencies around or above 99% across multiple methods, with detailed calibrations of energy loss (Δp/x) and Lorentz-angle effects, all in good agreement with Monte Carlo simulations. These results establish the SST as a well-understood subsystem ready for data-taking in collisions, with robust calibration, timing, and reconstruction performance that will underpin precision tracking and particle identification in CMS.

Abstract

During autumn 2008, the Silicon Strip Tracker was operated with the full CMS experiment in a comprehensive test, in the presence of the 3.8 T magnetic field produced by the CMS superconducting solenoid. Cosmic ray muons were detected in the muon chambers and used to trigger the readout of all CMS sub-detectors. About 15 million events with a muon in the tracker were collected. The efficiency of hit and track reconstruction were measured to be higher than 99% and consistent with expectations from Monte Carlo simulation. This article details the commissioning and performance of the Silicon Strip Tracker with cosmic ray muons.

Figures (17)

• Figure 1: Schematic cross section of the CMS tracker. Each line represents a detector module. Double lines indicate double-sided modules which deliver stereo hits.
• Figure 2: (left) Two APV25 data frames multiplexed, containing a time stamp and the sensor pulse height information. (right) A feature of the APV25 data stream, known as a tick mark, that is heavily used by the checkout and commissioning procedures. The left and right figures have sampling intervals of 25 ns and 1.04 ns, respectively.
• Figure 3: (Left) An example of the CR-RC pulse shape of a single APV25 chip, before and after the pulse shape tuning procedure. (Right) Pulse height measurements using the on-chip calibration circuitry of APV25 chips in the TEC+.
• Figure 4: (Left) Mean calibrated noise for individual APV25 chips on modules in the TOB single side layer 3. (Right) The ratio of minimum noise to median noise per APV25 chip. The distinct populations reflect the different noise sources within a module.
• Figure 5: (Left) Mean signal of leading strip in clusters associated to tracks as a function of the latency (25 ns steps), for each of the four partitions. (Right) Fine delay scan for the TOB layer 3, in deconvolution. The mean position (-14.2 ns) is including the mean time-of-flight of particles from the muon system to the silicon sensors (12 ns).