The Detector Control Power System of the Monitored Drift Tubesof the ATLAS Experiment

Theodoros AlexopoulosNTU Athens

TWEPP08September 17, 2008

Naxos, Greece

Outline

• Introductory Remarks about DCS

• MDT in ATLAS Experiment

• Structure of DCS

• Power Systems for MDT • Summary

Detector Control System (DCS)

DCS - Introduction

Detector Control Systems◦ Systems to control (via computers) the operation of a HEP

experiment◦ Appeared in 1980’s ◦ Increasing Complexity of the Experiments◦ Control of a large amount of parameters◦ Check of the smooth operation of the experiment

Characteristics◦ Operators don’t need to be sub-detector experts◦ Operation procedures become very easy ◦ Automatic Error Recovery procedures can be implemented

LAN-ETHERNET

OPIOperator Interface

OPIOperator Interface

IOCInput/Output Cntrl

IOCInput/Output Cntrl

Standard DCS Model(Client-Server Model)

OPI: Operator Interface , IOC: Input/Output Controller, IOS: Input/Output Servers

IOSInput/Output Server

IOSInput/Output Server

Clients

Servers

Controlle

rs

What does DCS involve?

Control, Configuration, Readout, Monitoring of Hardware devices (but not readout of event data)

Monitoring of external systems (LHC, Safety, Electrical, Gas, etc)

Implementation of Finite State Machine behavior

Communication with DAQLogging of data, status, storage in database

Partitioning

subDet 1 subDet 2 subDet 3

subSys 1 subSys 2

Driver 1 Driver 2

Device 1

central control

Control Units

Device Units

DCS Architecture

External systems

I/OControllers Analog/Digitalchannels,Fieldbuses

field

buse

s

SCA

DA

OPIterminal

Experiment Sub-Detectors &Experimental Equipment

LHCSafety Server

Beam

Se r

v er

Beam-lineLHC data

OPI

WAN

RemoteWorkstations

LocalWorkstations

I/O Serversdistributed inExperimental area O

PC

Typical DCS Architecture (physical subsystem layout)

bridge

brid

ge

OPI terminal

OPI terminal

LAN-ETHERNET

IOS

DBserver

Ora

cle

WAN

laptopOPI Remote Users

elmb caen-hv

Supervisory Control And Data Acquisition System: PVSS

Commercial product PVSSII from ETM, Austria LHC wide decision

Can be distributed over many stations

Flexible and open architecture

Basic functions for automatisation

Standardized interface to the hardware

Application programming interfaces

What is OPC? OPC defines a standard to interface programs (SCADA, HMI) and hardware devices in a control system.

OPC provides a multi-vendor interoperability.

No more specific drivers needed for each client

OPC (OLE for Process Control) is based on the MS object model COM/DCOM (Component Object Model). OPC includes 3 interface specifications:

Data Access Alarm and Event Handling Historical Data Access

Application X Application Y

Device A Device B

Application X Application Y

Device A Device B

OPC Interface OPC Interface

OPC Server OPC Server

OPC

10

LHC Experiments ATLAS

Diameter: 25 mLength: 44 mWeight: 7 kT ~100 M electronic chs1150 MDT chambers354 384 Tubes214 Tonnes725 m3 Gas Volume5520 m2 Chamber Area

A Toroidal LHC ApparatuS

ATLAS

Inner Detector CalorimetersMuons

TRT

SCT

Pixel LArg Tile

Trigger Precision

TGC RPC MDT CSC

46m

25mTRT: Transition Radiation DetectorSCT: Silicon Central TrackerTGC: Thin Gap ChamberRPC: Resistive Pad ChamberMDT: Monitored Drift TubesCSC: Cathode Strip Chamber

Muon Spectrometer

Two Basic parts Barrel και EndCap:◦ Barrel: 3 cylinders (inner-middle-

outer)◦ EndCap: 3 disks

Chambers: MDT, CSC, RPC, TGC

1150 MDT chambers354 384 Tubes214 Tones725 m3 Gas Volume5520 m3 Chamber Area

DCS Architecture in ATLAS

DCS_IS

Global Control Stations (GCS)

CERN

MagnetLHC

DSS

OperatorInterface

DataViewer Alarm Status Web

WAN

Subdetector Control Stations (SCS)

CIC Tile Pixel SCT TRT MDT TGC RPC CSC TDAQ LAr

Local Control Stations (LCS)

High/LowVoltage

Gas Alignment Temp B-field JTAG Rod Power

Front-End Systems

Finite States Machine (FSM)

ATLAS

MDT

Barrel A EndCap A

INNER MIDDLE OUTER

← GCS: Overall control of the detector

← SCS: Full local operation of sub-detector

← TTC Partitions →

BMS1C14 BML1C01

Barrel C EndCap C

← Layers

← Chambers

State & Status

Comm

ands

Structure of FSM:

Definition: A finite state machine (FSM) or finite state automaton or simply a state machine is a model of behavior composed of a finite number of states, transitions between those states and actions.

Pixel SCT TRT LAr Tile MDT TGC CSC RPCCIC

ATLAS

EMB-A EMB-B EMEC-A EMEC-C HEC FCAL

HV THERMO ROD FECHV-PS

Q2 Q3 Q4Q1

HV sector 1 HV sector 2 HV sector n

Global Control Station

Sub-detector Control Station

DAQ Partition

DAQDDC

Local Control Station

Geographic partition

Device Units

Operator Interface Finite State Machine (FSM) Architecture

The FSM (part of the JCOP Framework) is the main tool for the implementation of the full control hierarchy in ATLAS DCS

It is envisaged that the shift operators will operate DCS ONLY through the Operator Interface (based on the FSM) and the PVSS alarm screen

JCOP: Joint Controls Project

ATLASDCS

GCS

SCS 2 SCS N

TTCN.1

LCSN.1.1

DevN-1

LCSN.1.2

DevN

...

...

Shift Operator

TTC1.N

SCS 1

TTC1.1

TTC1.2

LCS1.1.1

LCS1.1.2

Sub-sys1.1.2.1

Sub-sys1.1.2.2

Dev1

Dev2

Dev4

Dev3

Dev5

...

Detector Expert

AnotherDetector Expert

Operator InterfaceFSM (Partitioning)

Operator InterfaceFSM (STATE and STATUS)

STATEREADY

NOT_READY

ON

STANDBY

RAMP_UP

….to be defined by the user

STATUSOK System works

fine

WARNING Low severity.System can goon working.

ALARM High severity.Serious error tobe fixed ASAP

FATAL Very high severity. The system cannot work.Co

mm

ands

STAT

US

TTC1.N

SCS 1

GCS

SCS 2 SCS N

TTC1.1

TTC1.2

TTCN.1

LCS1.1.1

LCS1.1.2

Sub-sys1.1.2.1

Sub-sys1.1.2.2

LCSN.1.1

LCSN.1.2

DevN-1

DevN

Dev1

Dev2

Dev4

Dev3

Dev5

...

...

...

Operator

Operator

STAT

E

Messages via a double Information Path – STATE & STATUS STATE defines the operational mode of the system (ON, OFF, etc) STATUS defines how well the system is working (OK, WARNING, ALARM, FATAL) Two parallel information paths. E.g. HV system is in RAMPING_UP state (which

takes several minutes) and an error triggers. The error is propagated through the STATUS while keeping the same STATE

ATLAS DCS at First Beam Events

CAEN Power Supplies modules

Crates

Boards LV: A3025 & A3016 HV: A3540P

MainframeCAEN SY1527

Branch ControllersA1676

19

Structure of MDT PSEltx Room

SY

1527

Sys

tem

mai

nfra

me

Bra

nch

Con

trolle

r

Experiment Area

Easy Module

Easy Module

Con

trol L

ogic

Con

trol L

ogic

Control Bus

48VPowerSource

Pow

er F

ail U

seIn

terr

upt

48VPowerSource

Pow

er L

ine

P. S

ervi

ce L

ine

48VBackupBattery

64 Crates

PS OPC server

EM EM

SY1527

OPC serverOPC

ClientOPC

Client

PS1

PS2 PS3 CSC

Run a single OPC server on PS1 pc (scattered system)

PS2 Barrel System

PS3 EndCap SystemEM

OPCClient

PVSS datapoints

Datapoints that correspond to hardware modules (channels, boards, crates, branch controllers, mainframe)

Datapoints that correspond to the chambers

Internal datapoints for the FSM for the OPC Server (communication with hardware) for the RDB manager (archiving of datapoint elements)

Alarms are activated for the datapoints and their elements

FSM trees

One FSM tree is implemented for each partition

Hierarchy is set for the parts of the system (nodes and their children)

Commands and states propagate correctly

FSM panels

FSM panels- EndCaps

Present status of PS system

Barrel

All chambers are incorporated in the system having LV modules

HV modules for all chambers in all sectors

Endcaps

Both Big Wheels and Small Wheels (one for each side) are incorporated to the system having both LV and HV modules

Summary

• ATLAS PS MDT DCS is a Robust System

• Is being used…

• Delivered & Tested on Time

Members of ATLAS NTU-Athens DCS group: Ex-membersT. Alexopoulos, T. Argyropoulos, E. Gazis, M. Bachtis, A. IliopoulosE. Mountricha, C. Tsarouchas, G. Tsipolitis