Sep 25 – 29, 2006
Valencia, Spain
Europe/Zurich timezone

The ALICE Silicon Pixel Detector Control system and online calibration tools

Sep 27, 2006, 12:20 PM
Valencia, Spain

Valencia, Spain

IFIC – Instituto de Fisica Corpuscular Edificio Institutos de Investgación Apartado de Correos 22085 E-46071 València SPAIN


Ivan Amos Cali (Universita degli Studi di Bari / CERN)


The ALICE Silicon Pixel Detector (SPD) contains nearly 10^7 hybrid pixel cells. About 2000 parameters and ~50000 DACs must be controlled in real-time during the detector integration, commissioning and operation. Information on each channel is stored in a configuration database. Timing and data management are critical issues. An overview of the SPD detector control system is presented, focusing on front-end controls and the SPD calibration strategy. An outlook of future implementations is presented.


The ALICE Silicon Pixel Detector (SPD) contains nearly 10^7 hybrid
pixel cells. About 2000 parameters must be controlled in real-time
during experimental data taking and in detector calibration runs.
Configuration and calibration for each channel (~50000
DACs in total) must be applied during the detector integration and commissioning as
well as at different phases operation. Information on each
channel is stored in a configuration database. Timing and data
management (~6 GB of raw data each calibration) are critical

--ALICE SPD layout--
The SPD constitutes the two innermost layers of the ALICE Inner
Tracking System (ITS). The SPD consists of 120 half-staves (HS),
each one consisting of two ladders (5 readout chips bump bonded to a
sensor for each ladder) and one Multi Chip Module (MCM). The MCM
distributes the timing signals, provides the required analog
references and controls the readout.
The communication between the MCM and the counting room is via optical links.
In the counting room 60 Link Receiver cards (LRx) perform zero
suppression and data encoding of the data streams. Three LRxs are plugged on 20 VME
based router module with optical links to the experiment DAQ and
trigger system.
The LV power is supplied by 20 CAEN A3009 modules (12 LV channel each)
and the sensor bias is provided by CAEN A1519 modules 10 (12 HV channel each).
The FE electronics power dissipation is ~1.3kW.
The cooling is based on an evaporative system with C4F10. The operational temperature
of the detector is ~25 °C.

--SPD Calibration procedure--
The detector calibration is based on electrical pulses generated on the FE
chips. The detector efficiency as well as the uniformity of
response of the pixels matrices are studied by varying the pulse
amplitude at various threshold settings (S-curves).
Automated configuration calibration procedures are implemented in
the Detector Control System (DCS) that is able to emulate the ALICE
DAQ and trigger system.
The raw data produced by the detector are analyzed online by the DCS
that provides a configuration parameter set that will be used by the
detector during the physics data taking.

--The SPD Control system--
The system is based on Supervisory Control And Data Acquisition
(SCADA), PVSS. The system controls two Front End Device (FED)
Servers (C++ based) that communicate with the hardware and carry out
the calibration and control procedures. PVSS agents control the
status and apply intervention as required by communicating both with
the Experiment Control System (ECS) and the FEDs. The detector data
are sent to a ROOT based analysis tool that produces online
information on the pixels response and define the configuration for
the whole system. The analysis tool is controlled via PVSS. The
system structure allows fast remote operator intervention and is
highly modular. The communication between the different software
platform applications is via Distributed Information Management
System (DIM - TCP/IP).
Most processes are fully automated in order to obtain the required reliability and
safety of operation.
The power system (HV and LV) and the cooling systems are controlled
directly by PVSS.
The logical control of the full system is carried out through a
Finite State Machine (FSM, SMI++ based) that defines the sequences
of commands to the different sub-systems. The FSM represents also the operator
interface to the detector and allows the interconnection with the
general ALICE DCS.

--SPD DCS status--
The prototype of the system is close to completion and has already
been used to test two SPD sectors during integration and
commissioning. The final electronic readout chain, the cooling and
power systems are operational. The integration with the ALICE DCS
and DAQ is in progress and will be soon completed. An overview of
the system status and performance, focusing on front-end controls
and analysis tools, is presented. The solutions for data management
and storage are discussed. An outlook of future work is shown.

Primary author

Ivan Amos Cali (Universita degli Studi di Bari / CERN)


Alexander Kluge (CERN) Cesar Torcato de Matos (CERN) Domenico Elia (Universita degli Studi di Bari) Gianluca Aglieri Rinella (CERN) Gianluca Guaglio (CERN) Giorgio Stefanini (CERN) Henrik Tydesjo (CERN) Marian Krivda (Slovak Academy of Sciences / CERN) Micheal Burns (CERN) Michel Morel (CERN) Michele Caselle (Universita degli Studi di Bari) Peter Chochula (CERN) Petra Riedler (CERN) Romualdo Santoro (Universita degli Studi di Bari) Simone Ceresa (Asssociazione Sviluppo Scientifico e Tecnologico de Piemonte (ASP)/CERN) Svetozar Kapusta (Comenius University / CERN) Vito Manzari (INFN di Bari)

Presentation materials