- Email: [email protected]
Construction and tests of the MRPC detectors for TOF in ALICE
ARTICLE IN PRESS Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Construction and tests of the MRPC detectors for TOF in ALICE A. Akindinov a, A. Alici b,c, P. Antonioli c, S. Arcelli b,c, Y.W. Baek d, M. Basile b,c, G. Cara Romeo c, L. Cifarelli b,c, F. Cindolo c, A. De Caro e, D. De Gruttola e,, S. De Pasquale e, M. Fusco Girard e, C. Guarnaccia e, D. Hatzifotiadou c, H.T. Jung d, W.W. Jung d, D.S. Kim d, D.W. Kim d, H.N. Kim d, J.S. Kim d, S. Kiselev a, G. Laurenti c, K. Lee d, S.C. Lee d, M.L. Luvisetto c, D. Malkevich a, A. Margotti c, R. Nania c, A. Nedosekin a, F. Noferini c,f, P. Pagano e, A. Pesci c, R. Preghenella b,c, G. Russo e, M. Ryabinin a, E. Scapparone c, G. Scioli b,c, A. Silenzi b,c, M. Tchoumakov a, K. Voloshin a, M.C.S. Williams c, B. Zagreev a, C. Zampolli c,f, A. Zichichi b,c,f a
Institute for Theoretical and Experimental Physics, Moscow, Russia `, Bologna, Italy Dipartimento di Fisica dell’Universita c Sezione INFN, Bologna, Italy d Department of Physics, Kangnung National University, Kangnung, Republic of Korea e ` and INFN, Salerno, Italy Dipartimento di Fisica dell’Universita f Museo Storico della Fisica e Centro Studi e Ricerche ‘‘Enrico Fermi’’, Roma, Italy b
a r t i c l e in f o
a b s t r a c t
Available online 24 December 2008
CERN-LHC (Large Hadron Collider) accelerator facility will provide heavy ions (Pb–Pb) collisions with a center-of-mass (CM) energy of about 5.5 TeV per nucleon pair. In the extreme conditions of temperature and energy density created in such collisions, a transition from hadronic matter towards a deconfined state of quarks and gluons is predicted by Quantum Chromodynamics (QCD) calculations on the lattice. The Time Of Flight (TOF) detector system of the ALICE (A Large Ion Collider Experiment) apparatus, is presently progressing in the assembling process at LHC at CERN. The TOF, in combination with the other central tracking detectors of ALICE provides an excellent Particle IDentification (PID) in the momentum range 0:222:5 GeV=c for K=p and up to 4 GeV/c for K/p. The ALICE TOF is a barrel detector consisting of double-stack Multigap Resistive Plate Chamber (MRPC) strips, equipped with readout pads. The MRPC is characterized by an intrinsic time resolution below 50 ps and an efficiency over 99%. The assembling procedures, the tests of mechanics, cooling system and electronics of the 8 m long TOF ‘‘supermodules’’, together with the performance tests before installation in the experimental area, will be presented. & 2008 Elsevier B.V. All rights reserved.
Keywords: ALICE MRPC Strip Module Supermodule Front-End ASIC (FEA) FEA controller (FEAC)
1. Introduction In ultra-relativistic heavy-ion collisions, temperature and energy densities are expected to be large enough to lead to the transition from hadronic matter to a deconfined state of quarks and gluons predicted by QCD [1]. The results from the first fixed target experiments at the CERN SPS have given a first indication that a weakly interacting system of quarks and gluons, like an ideal gas, is formed during the collision [2]. Following the experimental observations at the RHIC collider, a strongly interacting system of quarks and gluons, like an ideal liquid, could be created during the collisions [3]. The study of the deconfined
Corresponding author.
E-mail address: [email protected] (D. De Gruttola). 0168-9002/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2008.12.075
phase of the matter can be pursued with LHC, where the available energy in the collision will be a factor 30 greater than the one at RHIC. The ALICE (A Large Ion Collider Experiment) [4] is the CERN-LHC experiment specifically designed for the study of heavy ion collisions. The Time Of Flight (TOF) detector system is a largearea array of gaseous detectors that covers the central pseudorapidity region ðjZjp0:9Þ. It is devoted to PID in the intermediate momentum range (few hundred MeV/c to few GeV/c). It has a modular structure corresponding to 18 sectors in the azimuthal angle j ð0 pjp360 Þ and to five segments in the longitudinal coordinate along the beam axis z. The whole device is located inside a cylindrical shell with an internal radius of 370 cm and an external radius of 399 cm. The basic detector of the TOF system is the MRPC (Multigap Resistive Plate Chamber) ‘‘strip’’ [5] (a crosssection of a MRPC strip in Fig. 1), providing excellent time resolution (below 50 ps) and efficiency (above 99%). With this
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
Fig. 1. Cross-section of ALICE TOF MRPC strip.
Fig. 2. A scheme and a picture of tilted geometry for TOF MRPC strips, in an intermediate module.
659
ARTICLE IN PRESS 660
A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
potential, the PID capabilities of the TOF allows a 3s separation in the range 0:222:5 GeV=c for k/p and up to 4 GeV/c for k/p, using the TOF formula sffiffiffiffiffiffiffiffiffiffiffiffiffi t2 m¼p 2 1 l
where m is the mass of the particle, p the momentum, l the track length and t the TOF measurement.
2. Modules assembling and tests
Fig. 3. One of the 18 supermodules of the ALICE TOF system.
The configuration of the MRPC chosen for the TOF system of ALICE is the ‘‘double-stack’’, characterized by an internal anode and two external cathodes; inside each stack there is a series of gas gaps, bounded by glass plates, equally spaced using a nylon wire as spacer. The high voltage (HV) is applied through electrodes connected to the external surfaces of the outer glasses of the stack of resistive plates. All the internal plates are not connected to the HV and are electrically floating. Each gap has a width of 250 mm and is filled with a gas mixture C2 F4 H2 (90%)–C4 H10 (5%)–SF6 (5%). The signal of a MRPC is picked up by 96 pads placed on the anode/cathode PCBs (Fig. 1). The MRPC strips are housed in special gas tight boxes called ‘‘modules’’. The modules are assembled together in group of five to compose a ‘‘supermodule’’ (see Section 3 and Fig. 4). To cover the barrel region of ALICE, 90 modules, arranged in 18 supermodules are needed. There are three different types of modules, containing a different number of MRPC strips: the central one with fifteen strips inside, the intermediate and the external modules with nineteen strips. A particular tilted strip geometry inside the module has been chosen to minimize the impact angle of the track (Fig. 2). The assembling of the 90 modules, needed for the entire TOF, has been performed in the INFN laboratories of Bologna (Italy); each module consists of vetronite tank filled with the strips an with the gas mixture, closed by an aluminium cover equipped with PCBs, which ensure the interface between the MRPC pads and the Front-End ASIC (FEA) mounted on the PCBs (Fig. 4).
Fig. 4. Pulse injection line on the PCB of the strip and FEAs connected to the module.
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
661
Entires 311 Mean 0.9648 RMS 0.02948 Overflow 0
channels / 0.1
50 Real Data
40
Simulation
30 20 10 0 0.4
0.5
0.6
0.7
0.8 0.9 1 efficiency
1.1
1.2
1.3
1.4
Fig. 6. Real data and simulation from the cosmic rays facility.
(TDC) of the readout electronics is checked. At this phase of the construction, it is still possible to recover the possible dead channels, opening the module and checking the problem directly on the strip or on the connections with the interface-card.
3. Supermodules assembling and tests
Fig. 5. Cosmic rays facility at CERN.
A complex detector like ALICE TOF with about 160 000 electronic channels, covering a total sensitive area 150 m2 requires a careful construction procedure and a set of tests. Some tests are performed on the MRPC strips after their assembling in the modules: HV connections, gap size, glasses resistivity and cosmic rays tests. The tests performed on the modules are: gas tightness to prevent gas leakage from the modules, HV connections in air to avoid spark near the active area of the strips and pulser to find possible dead channels. Gas tightness: This test is performed several times: the first time after the installation of the strips in the module. A further test is performed after the transport of the module to CERN. When the modules are assembled together to make the supermodule (Fig. 3), the gas tightness is performed again, to be sure that the installation of the mechanical pieces had not damaged the modules. The limit of the accepted leakage is 1 mb/h. High voltage: The strips inside the module are connected to the HV distributors in group of five or four (it depends of the number of strips inside the module). A preliminary HV test is performed on the strips before the insertion in the module. On the other end the module without strips is also tested, in order to check the quality of the connections; this operation is made in air at HV ¼ 8 kV. When the strips are inserted, a new test is performed at HV ¼ 3 kV. The accepted value for dark current is a few mA. Pulser: In Fig. 4 the location of the pulse injection line is indicated on the PCB of the strip; this line has been foreseen to send a pulse on each readout pad and check the readout signal quality; the test is very simple: a pulse signal is induced to the pads and the corresponding hit on the Time to Digital Converters
After having passed the tests, the module is carried from the INFN laboratories to the CERN laboratories, where the assembling of the 18 supermodules of the TOF system is carried on. Before assembling the supermodule, each module is subjected to a severe performance test on the cosmic rays facility installed in the CERN hall 167 (Fig. 5). The data collected are compared with the full simulation data which take in account the cosmic rays flux, the trigger system (two scintillators planes) and the geometry of the strips in the module. In Fig. 6 a comparison between the real data and the simulation is shown. The frontend and the readout electronics on this facility is the same as the one used in the experiment. The five modules coming from the cosmic rays facility are placed on a table and aligned and the mechanical pieces for the rigidity of the structure are mounted on the modules; after this operation, a new gas tightness test is performed in order to avoid leakage due to the mechanical mounting. After the mechanical pieces, the cooling pipes and the readout crates are mounted on the modules. A pipe tightness test of the FEA cooling system and of the crates cooling system starts using first compressed air at 100 mbar and then water at 10 bar for 20 min. The next step is the cabling of the supermodule: after having put the pulser cables, the HV and LV cables, the signal cables (Amphenol skewclear cables1) are placed to connect the FEA (previously connected to the interface-card of the modules) to the TDC readout Modules in the crates. The readout crates are kept working and stressed for a whole week; they are subjected to cooling test, a check of the LV channels and many quality tests of the boards inserted in the crates. The readout modules have been specially designed for the TOF system; they consist of:
Data Readout Module (DRM), performing the following operations: VME master, TDC Readout Module (TRM) readout, DAQ via Detector Data Link (DDL), slow control and Level0 (L0), Level1 (L1) and Level2 (L2) trigger via Timing Trigger Control (TTC). 1 Designed to reduce electromagnetic interference, due to tight control of within-pair skew.
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
1
1
0.8
0.8 efficiency
efficiency
662
0.6 0.4 0.2
0.4 0.2
500
1000 1500 channel Crate 0
0
2000
1
1
0.8
0.8 efficiency
efficiency
0
0.6
0.6 0.4 0.2 0
500
1000 1500 channel Crate 1
2000
500
1000 1500 channel Crate 3
2000
0.6 0.4 0.2
500
1000 1500 channel Crate 2
2000
0
Fig. 7. Results of the pulser test on nine supermodules; 0.1% is the dead channels rate.
Fig. 8. FEA and the service card FEAC.
Clock and Pulser Distribution Module (CPDM), devoted to
ALICE clock distribution and pulser to the MRPC pads. Local Trigger Module (LTM): FEA LV monitoring, FEA threshold setting and monitoring, temperature monitoring and FEA OR signal transmission to Cosmic and Topology Trigger Module (CTTM). TDC Readout Module (TRM), that hosts the TDC chips (30 per each TRM, each TDC reading 8 MRPC channels).
A pulser test similar to the one made on the single modules is performed in order to check dead channels after the supermodule assembling; in Fig. 7 it is possible to note the very low percentage of dead channels in the TOF system (0.1%). A test with pulser off is also performed in order to find possible noisy channels; after having tested 13 supermodules the fraction of noisy and dead channels is few channels over more than one hundred thousand. The TOF will contribute to the ALICE first level (L1) trigger
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
663
Fig. 9. Some supermodules installed in the magnet.
decision, taking the OR signals of couples of daisy-chained FEAs and, for this reason, the trigger chain2 is tested looking at the OR signals of the TDCs during a pulser run.
4. Front-end electronics Each strip is directly connected to four FEAs, each one containing three NINO ASIC chips [6]; a NINO ASIC has the following features:
it reads 8 channels; it has a LVDS (Low Voltage Differential Signal) input/output; it has a fast amplifier to minimize time jitter; the threshold of the discriminator is adjustable in the range 10–100 fC.
The FEAs are arranged in daisy-chain; a group of 10 or 123 FEAs is connected to a FEAC, which plays the role of FEA service card. FEACs have an important task in the supermodule set-up; they provide the voltage to power on the FEAs and allow to communicate with them, in order to set the threshold voltages on the discriminator and collect the OR signal from the channels of the daisy-chained FEAs (this is an important signal for the trigger of the experiment); besides, they can monitor the temperature of different areas of the detector, by a temperature sensor inserted in the circuit. In Fig. 8 a FEA and a FEAC are shown. In the laboratories of University of Salerno (Italy), some tests on the FEACs are performed: circuit lines, transmission of the LVDS and temperature and calibration curve.
Circuit lines and LVDS: the first one is useful to check open and short circuits in the circuit lines of FEAC, in order to ensure a good power supply working (test on ground and LV lines) and the right working of OR signals transmission line and thresholds setting lines. LVDS is the kind of differential signal outcoming from FEAs; the test is carried out sending to the FEAC a LVDS signal generated with a FPGA (Field Program2 3
LTM, trigger cables and FEA Controller (FEAC). The number depends on the position of the FEA in the supermodule.
mable Gate Array) and checking the absence of distortions, reductions of the amplitude and crosstalk along the circuit lines; Temperature: the FEAC has a low-voltage analog temperature sensor, that works in a range of about 50 C. The test is performed reading the temperature sensor voltage, after having put the FEAC at three different values of temperature: 5, 25 and 35 C. With the achieved data it is possible to check the proper functioning of cards at different temperature values and to get a calibration curve. This test is very important to have a good temperature monitoring system inside the supermodules, whose electronic cards will work hard for a lot of years.
All these tests are performed with a Labview application in order to make the tests easier and faster.
5. Installation and commissioning In Fig. 9 the ALICE magnet with some supermodules installed is shown. The installation of a supermodule is a complex and delicate operation; it is performed with the help of mechanical structures specially designed to bring the supermodule, to rotate it and to achieve the right angle, in order to install the supermodules in the 18 different sectors of the TOF frame. After the installation, the commissioning stage starts, i.e. final cabling, cooling, gas tightness tests, power on and readout tests. The cabling operation in the magnet consists in:
plugging HV cables coming from ALICE services system in the supermodule distributors;
plugging power cables to provide 48 V to switch the crates on; plugging brunch control cables, used to monitor and power on all the LV channels of the crates;
connecting gas ALICE pipes to the supermodules connectors; connecting cooling pipes to the supermodules connectors. After each operation some tests on the connected cables and pipes are performed in order to complete the commissioning
ARTICLE IN PRESS 664
A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
phase. After the first commissioning phase of the first supermodules installed in the magnet, a cosmic rays run has been performed in December 2007; a System Control And Data Acquisition (SCADA) software package chosen by CERN as standard for control systems of LHC experiments—PVSSII, for the slow control has been used during this run and it will be used during the next runs and the experiment. Some cosmic muons has been detected and integration runs with the other sub-detectors of ALICE, as V0, T0, ITS, TRD, Muon tracker and ACORDE, was successfully performed.
center-of-mass energy of about 5.5 TeV per nucleon pair. The assembling of the supermodules has finished with the very good result of few dead and noisy electronics channels, over about 100 thousand. All the supermodules will be installed in the ALICE magnet for the end of April 2008. A first cosmic run has been performed in December 2007, with good results concerning the muon detection and the integration with the other ALICE subdetectors.
References 6. Conclusions A detector with efficiency above 99% and a time resolution below 50 ps has been chosen to build the TOF system of ALICE, in order to perform PID in the momentum range 0:222:5 GeV=c for K=p and up to 4 GeV/c for K/p; this system is divided in 18 sectors (supermodules) in the azimuthal angle j ð0 pjp360 Þ and in five segments in the longitudinal coordinate along the beam axis z of LHC, that will provide collisions of heavy ions (Pb–Pb) with a
[1] E. Shuryak, The QCD vacuum, chiral phase transition and Quark-Gluon Plasma, arXiv:hep-ph/9503427v1. [2] U. Heinz, M. Jacob, Evidence for a new state of matter: an assessment of the results from the CERN lead beam programme, arXiv:nucl-th/0002042v1. [3] M.H. Thoma, Complex plasmas as a model for the Quark-Gluon-Plasma liquid, arXiv:hep-ph/0509154v1. [4] ALICE Collaboration, ALICE: Physics Performance Report, volume I, J. Phys., G30, 2004, 1517–1763 SLAC JPHGB, G30, 151. [5] E. Cerron Zeballos, et al., Nucl. Instr. and Meth. A 374 (1996) 132 SLAC NUIMA, A374, 132. [6] F. Anghinolfi, et al., Nucl. Instr. and Meth. A 533 (2004) 183.
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Construction and tests of the MRPC detectors for TOF in ALICE A. Akindinov a, A. Alici b,c, P. Antonioli c, S. Arcelli b,c, Y.W. Baek d, M. Basile b,c, G. Cara Romeo c, L. Cifarelli b,c, F. Cindolo c, A. De Caro e, D. De Gruttola e,, S. De Pasquale e, M. Fusco Girard e, C. Guarnaccia e, D. Hatzifotiadou c, H.T. Jung d, W.W. Jung d, D.S. Kim d, D.W. Kim d, H.N. Kim d, J.S. Kim d, S. Kiselev a, G. Laurenti c, K. Lee d, S.C. Lee d, M.L. Luvisetto c, D. Malkevich a, A. Margotti c, R. Nania c, A. Nedosekin a, F. Noferini c,f, P. Pagano e, A. Pesci c, R. Preghenella b,c, G. Russo e, M. Ryabinin a, E. Scapparone c, G. Scioli b,c, A. Silenzi b,c, M. Tchoumakov a, K. Voloshin a, M.C.S. Williams c, B. Zagreev a, C. Zampolli c,f, A. Zichichi b,c,f a
Institute for Theoretical and Experimental Physics, Moscow, Russia `, Bologna, Italy Dipartimento di Fisica dell’Universita c Sezione INFN, Bologna, Italy d Department of Physics, Kangnung National University, Kangnung, Republic of Korea e ` and INFN, Salerno, Italy Dipartimento di Fisica dell’Universita f Museo Storico della Fisica e Centro Studi e Ricerche ‘‘Enrico Fermi’’, Roma, Italy b
a r t i c l e in f o
a b s t r a c t
Available online 24 December 2008
CERN-LHC (Large Hadron Collider) accelerator facility will provide heavy ions (Pb–Pb) collisions with a center-of-mass (CM) energy of about 5.5 TeV per nucleon pair. In the extreme conditions of temperature and energy density created in such collisions, a transition from hadronic matter towards a deconfined state of quarks and gluons is predicted by Quantum Chromodynamics (QCD) calculations on the lattice. The Time Of Flight (TOF) detector system of the ALICE (A Large Ion Collider Experiment) apparatus, is presently progressing in the assembling process at LHC at CERN. The TOF, in combination with the other central tracking detectors of ALICE provides an excellent Particle IDentification (PID) in the momentum range 0:222:5 GeV=c for K=p and up to 4 GeV/c for K/p. The ALICE TOF is a barrel detector consisting of double-stack Multigap Resistive Plate Chamber (MRPC) strips, equipped with readout pads. The MRPC is characterized by an intrinsic time resolution below 50 ps and an efficiency over 99%. The assembling procedures, the tests of mechanics, cooling system and electronics of the 8 m long TOF ‘‘supermodules’’, together with the performance tests before installation in the experimental area, will be presented. & 2008 Elsevier B.V. All rights reserved.
Keywords: ALICE MRPC Strip Module Supermodule Front-End ASIC (FEA) FEA controller (FEAC)
1. Introduction In ultra-relativistic heavy-ion collisions, temperature and energy densities are expected to be large enough to lead to the transition from hadronic matter to a deconfined state of quarks and gluons predicted by QCD [1]. The results from the first fixed target experiments at the CERN SPS have given a first indication that a weakly interacting system of quarks and gluons, like an ideal gas, is formed during the collision [2]. Following the experimental observations at the RHIC collider, a strongly interacting system of quarks and gluons, like an ideal liquid, could be created during the collisions [3]. The study of the deconfined
Corresponding author.
E-mail address: [email protected] (D. De Gruttola). 0168-9002/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2008.12.075
phase of the matter can be pursued with LHC, where the available energy in the collision will be a factor 30 greater than the one at RHIC. The ALICE (A Large Ion Collider Experiment) [4] is the CERN-LHC experiment specifically designed for the study of heavy ion collisions. The Time Of Flight (TOF) detector system is a largearea array of gaseous detectors that covers the central pseudorapidity region ðjZjp0:9Þ. It is devoted to PID in the intermediate momentum range (few hundred MeV/c to few GeV/c). It has a modular structure corresponding to 18 sectors in the azimuthal angle j ð0 pjp360 Þ and to five segments in the longitudinal coordinate along the beam axis z. The whole device is located inside a cylindrical shell with an internal radius of 370 cm and an external radius of 399 cm. The basic detector of the TOF system is the MRPC (Multigap Resistive Plate Chamber) ‘‘strip’’ [5] (a crosssection of a MRPC strip in Fig. 1), providing excellent time resolution (below 50 ps) and efficiency (above 99%). With this
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
Fig. 1. Cross-section of ALICE TOF MRPC strip.
Fig. 2. A scheme and a picture of tilted geometry for TOF MRPC strips, in an intermediate module.
659
ARTICLE IN PRESS 660
A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
potential, the PID capabilities of the TOF allows a 3s separation in the range 0:222:5 GeV=c for k/p and up to 4 GeV/c for k/p, using the TOF formula sffiffiffiffiffiffiffiffiffiffiffiffiffi t2 m¼p 2 1 l
where m is the mass of the particle, p the momentum, l the track length and t the TOF measurement.
2. Modules assembling and tests
Fig. 3. One of the 18 supermodules of the ALICE TOF system.
The configuration of the MRPC chosen for the TOF system of ALICE is the ‘‘double-stack’’, characterized by an internal anode and two external cathodes; inside each stack there is a series of gas gaps, bounded by glass plates, equally spaced using a nylon wire as spacer. The high voltage (HV) is applied through electrodes connected to the external surfaces of the outer glasses of the stack of resistive plates. All the internal plates are not connected to the HV and are electrically floating. Each gap has a width of 250 mm and is filled with a gas mixture C2 F4 H2 (90%)–C4 H10 (5%)–SF6 (5%). The signal of a MRPC is picked up by 96 pads placed on the anode/cathode PCBs (Fig. 1). The MRPC strips are housed in special gas tight boxes called ‘‘modules’’. The modules are assembled together in group of five to compose a ‘‘supermodule’’ (see Section 3 and Fig. 4). To cover the barrel region of ALICE, 90 modules, arranged in 18 supermodules are needed. There are three different types of modules, containing a different number of MRPC strips: the central one with fifteen strips inside, the intermediate and the external modules with nineteen strips. A particular tilted strip geometry inside the module has been chosen to minimize the impact angle of the track (Fig. 2). The assembling of the 90 modules, needed for the entire TOF, has been performed in the INFN laboratories of Bologna (Italy); each module consists of vetronite tank filled with the strips an with the gas mixture, closed by an aluminium cover equipped with PCBs, which ensure the interface between the MRPC pads and the Front-End ASIC (FEA) mounted on the PCBs (Fig. 4).
Fig. 4. Pulse injection line on the PCB of the strip and FEAs connected to the module.
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
661
Entires 311 Mean 0.9648 RMS 0.02948 Overflow 0
channels / 0.1
50 Real Data
40
Simulation
30 20 10 0 0.4
0.5
0.6
0.7
0.8 0.9 1 efficiency
1.1
1.2
1.3
1.4
Fig. 6. Real data and simulation from the cosmic rays facility.
(TDC) of the readout electronics is checked. At this phase of the construction, it is still possible to recover the possible dead channels, opening the module and checking the problem directly on the strip or on the connections with the interface-card.
3. Supermodules assembling and tests
Fig. 5. Cosmic rays facility at CERN.
A complex detector like ALICE TOF with about 160 000 electronic channels, covering a total sensitive area 150 m2 requires a careful construction procedure and a set of tests. Some tests are performed on the MRPC strips after their assembling in the modules: HV connections, gap size, glasses resistivity and cosmic rays tests. The tests performed on the modules are: gas tightness to prevent gas leakage from the modules, HV connections in air to avoid spark near the active area of the strips and pulser to find possible dead channels. Gas tightness: This test is performed several times: the first time after the installation of the strips in the module. A further test is performed after the transport of the module to CERN. When the modules are assembled together to make the supermodule (Fig. 3), the gas tightness is performed again, to be sure that the installation of the mechanical pieces had not damaged the modules. The limit of the accepted leakage is 1 mb/h. High voltage: The strips inside the module are connected to the HV distributors in group of five or four (it depends of the number of strips inside the module). A preliminary HV test is performed on the strips before the insertion in the module. On the other end the module without strips is also tested, in order to check the quality of the connections; this operation is made in air at HV ¼ 8 kV. When the strips are inserted, a new test is performed at HV ¼ 3 kV. The accepted value for dark current is a few mA. Pulser: In Fig. 4 the location of the pulse injection line is indicated on the PCB of the strip; this line has been foreseen to send a pulse on each readout pad and check the readout signal quality; the test is very simple: a pulse signal is induced to the pads and the corresponding hit on the Time to Digital Converters
After having passed the tests, the module is carried from the INFN laboratories to the CERN laboratories, where the assembling of the 18 supermodules of the TOF system is carried on. Before assembling the supermodule, each module is subjected to a severe performance test on the cosmic rays facility installed in the CERN hall 167 (Fig. 5). The data collected are compared with the full simulation data which take in account the cosmic rays flux, the trigger system (two scintillators planes) and the geometry of the strips in the module. In Fig. 6 a comparison between the real data and the simulation is shown. The frontend and the readout electronics on this facility is the same as the one used in the experiment. The five modules coming from the cosmic rays facility are placed on a table and aligned and the mechanical pieces for the rigidity of the structure are mounted on the modules; after this operation, a new gas tightness test is performed in order to avoid leakage due to the mechanical mounting. After the mechanical pieces, the cooling pipes and the readout crates are mounted on the modules. A pipe tightness test of the FEA cooling system and of the crates cooling system starts using first compressed air at 100 mbar and then water at 10 bar for 20 min. The next step is the cabling of the supermodule: after having put the pulser cables, the HV and LV cables, the signal cables (Amphenol skewclear cables1) are placed to connect the FEA (previously connected to the interface-card of the modules) to the TDC readout Modules in the crates. The readout crates are kept working and stressed for a whole week; they are subjected to cooling test, a check of the LV channels and many quality tests of the boards inserted in the crates. The readout modules have been specially designed for the TOF system; they consist of:
Data Readout Module (DRM), performing the following operations: VME master, TDC Readout Module (TRM) readout, DAQ via Detector Data Link (DDL), slow control and Level0 (L0), Level1 (L1) and Level2 (L2) trigger via Timing Trigger Control (TTC). 1 Designed to reduce electromagnetic interference, due to tight control of within-pair skew.
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
1
1
0.8
0.8 efficiency
efficiency
662
0.6 0.4 0.2
0.4 0.2
500
1000 1500 channel Crate 0
0
2000
1
1
0.8
0.8 efficiency
efficiency
0
0.6
0.6 0.4 0.2 0
500
1000 1500 channel Crate 1
2000
500
1000 1500 channel Crate 3
2000
0.6 0.4 0.2
500
1000 1500 channel Crate 2
2000
0
Fig. 7. Results of the pulser test on nine supermodules; 0.1% is the dead channels rate.
Fig. 8. FEA and the service card FEAC.
Clock and Pulser Distribution Module (CPDM), devoted to
ALICE clock distribution and pulser to the MRPC pads. Local Trigger Module (LTM): FEA LV monitoring, FEA threshold setting and monitoring, temperature monitoring and FEA OR signal transmission to Cosmic and Topology Trigger Module (CTTM). TDC Readout Module (TRM), that hosts the TDC chips (30 per each TRM, each TDC reading 8 MRPC channels).
A pulser test similar to the one made on the single modules is performed in order to check dead channels after the supermodule assembling; in Fig. 7 it is possible to note the very low percentage of dead channels in the TOF system (0.1%). A test with pulser off is also performed in order to find possible noisy channels; after having tested 13 supermodules the fraction of noisy and dead channels is few channels over more than one hundred thousand. The TOF will contribute to the ALICE first level (L1) trigger
ARTICLE IN PRESS A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
663
Fig. 9. Some supermodules installed in the magnet.
decision, taking the OR signals of couples of daisy-chained FEAs and, for this reason, the trigger chain2 is tested looking at the OR signals of the TDCs during a pulser run.
4. Front-end electronics Each strip is directly connected to four FEAs, each one containing three NINO ASIC chips [6]; a NINO ASIC has the following features:
it reads 8 channels; it has a LVDS (Low Voltage Differential Signal) input/output; it has a fast amplifier to minimize time jitter; the threshold of the discriminator is adjustable in the range 10–100 fC.
The FEAs are arranged in daisy-chain; a group of 10 or 123 FEAs is connected to a FEAC, which plays the role of FEA service card. FEACs have an important task in the supermodule set-up; they provide the voltage to power on the FEAs and allow to communicate with them, in order to set the threshold voltages on the discriminator and collect the OR signal from the channels of the daisy-chained FEAs (this is an important signal for the trigger of the experiment); besides, they can monitor the temperature of different areas of the detector, by a temperature sensor inserted in the circuit. In Fig. 8 a FEA and a FEAC are shown. In the laboratories of University of Salerno (Italy), some tests on the FEACs are performed: circuit lines, transmission of the LVDS and temperature and calibration curve.
Circuit lines and LVDS: the first one is useful to check open and short circuits in the circuit lines of FEAC, in order to ensure a good power supply working (test on ground and LV lines) and the right working of OR signals transmission line and thresholds setting lines. LVDS is the kind of differential signal outcoming from FEAs; the test is carried out sending to the FEAC a LVDS signal generated with a FPGA (Field Program2 3
LTM, trigger cables and FEA Controller (FEAC). The number depends on the position of the FEA in the supermodule.
mable Gate Array) and checking the absence of distortions, reductions of the amplitude and crosstalk along the circuit lines; Temperature: the FEAC has a low-voltage analog temperature sensor, that works in a range of about 50 C. The test is performed reading the temperature sensor voltage, after having put the FEAC at three different values of temperature: 5, 25 and 35 C. With the achieved data it is possible to check the proper functioning of cards at different temperature values and to get a calibration curve. This test is very important to have a good temperature monitoring system inside the supermodules, whose electronic cards will work hard for a lot of years.
All these tests are performed with a Labview application in order to make the tests easier and faster.
5. Installation and commissioning In Fig. 9 the ALICE magnet with some supermodules installed is shown. The installation of a supermodule is a complex and delicate operation; it is performed with the help of mechanical structures specially designed to bring the supermodule, to rotate it and to achieve the right angle, in order to install the supermodules in the 18 different sectors of the TOF frame. After the installation, the commissioning stage starts, i.e. final cabling, cooling, gas tightness tests, power on and readout tests. The cabling operation in the magnet consists in:
plugging HV cables coming from ALICE services system in the supermodule distributors;
plugging power cables to provide 48 V to switch the crates on; plugging brunch control cables, used to monitor and power on all the LV channels of the crates;
connecting gas ALICE pipes to the supermodules connectors; connecting cooling pipes to the supermodules connectors. After each operation some tests on the connected cables and pipes are performed in order to complete the commissioning
ARTICLE IN PRESS 664
A. Akindinov et al. / Nuclear Instruments and Methods in Physics Research A 602 (2009) 658–664
phase. After the first commissioning phase of the first supermodules installed in the magnet, a cosmic rays run has been performed in December 2007; a System Control And Data Acquisition (SCADA) software package chosen by CERN as standard for control systems of LHC experiments—PVSSII, for the slow control has been used during this run and it will be used during the next runs and the experiment. Some cosmic muons has been detected and integration runs with the other sub-detectors of ALICE, as V0, T0, ITS, TRD, Muon tracker and ACORDE, was successfully performed.
center-of-mass energy of about 5.5 TeV per nucleon pair. The assembling of the supermodules has finished with the very good result of few dead and noisy electronics channels, over about 100 thousand. All the supermodules will be installed in the ALICE magnet for the end of April 2008. A first cosmic run has been performed in December 2007, with good results concerning the muon detection and the integration with the other ALICE subdetectors.
References 6. Conclusions A detector with efficiency above 99% and a time resolution below 50 ps has been chosen to build the TOF system of ALICE, in order to perform PID in the momentum range 0:222:5 GeV=c for K=p and up to 4 GeV/c for K/p; this system is divided in 18 sectors (supermodules) in the azimuthal angle j ð0 pjp360 Þ and in five segments in the longitudinal coordinate along the beam axis z of LHC, that will provide collisions of heavy ions (Pb–Pb) with a
[1] E. Shuryak, The QCD vacuum, chiral phase transition and Quark-Gluon Plasma, arXiv:hep-ph/9503427v1. [2] U. Heinz, M. Jacob, Evidence for a new state of matter: an assessment of the results from the CERN lead beam programme, arXiv:nucl-th/0002042v1. [3] M.H. Thoma, Complex plasmas as a model for the Quark-Gluon-Plasma liquid, arXiv:hep-ph/0509154v1. [4] ALICE Collaboration, ALICE: Physics Performance Report, volume I, J. Phys., G30, 2004, 1517–1763 SLAC JPHGB, G30, 151. [5] E. Cerron Zeballos, et al., Nucl. Instr. and Meth. A 374 (1996) 132 SLAC NUIMA, A374, 132. [6] F. Anghinolfi, et al., Nucl. Instr. and Meth. A 533 (2004) 183.