color=blue>current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
Resistive plate chamber performance in the BaBar IFR system
100
Resistive plate chamber performance in the BaBar
IFR system
F. Anulli, S. Bagnasco, R. Baldini, H.R. Band, R. Bionta, J.E. Brau, V. Brigljevic, A. Buzzo,
A. Calcaterra, M. Carpinelli, C. Cartaro, N. Cavallo, G. Crosetti, R. de Sangro, G. De Nardo,
A. Eichenbaum, F. Fabozzi, D. Falciai, F. Ferrarotto, F. Ferroni, G. Finocchiaro, F. Forti, R. Frey,
C. Gatto, E. Grauges, M. Iwasaki, J.R. Johnson, D.J. Lange, L. Lista, M. Lo Vetere, C. Lu,
M. Macri, R. Messner, T.B. Moore, S. Morganti, N. Neri, H. Neal, A. Palano, E. Paoloni,
P. Paolucci, S. Passaggio, F. Pastore, P. Patteri, I. Peruzzi, M. Piccolo, D. Piccolo, G. Piredda,
E. Robutti, A. Roodman, A. Santroni, C. Sciacca, N.B. Sinev, D. Strom, A. Soha, S. Tosi, J. Vavra,
W.J. Wisniewski, D.M. Wright, Y. Xie, A. Zallo
Abstract
The BaBar Collaboration has operated a system covering
over 2000 m
2
with resistive plate chambers for nearly three
years. The chambers are constructed of bakelite sheets sep-
arated by 2 mm. The inner surfaces are coated with linseed
oil. This system provides muon and neutral hadron detec-
tion for BaBar. Installation and commissioning were com-
pleted in 1998, and operation began mid-year 1999. While
initial performance of the system reached design, over time,
a signicant fraction of the RPCs demonstrated signicant
degradation, marked by increased currents and reduced e-
ciency. A coordinated eort of investigations have identied
many of the elements responsible for the degradation.
Manuscript received November 23, 2001; revised February 3, 2002.
This work was supported by the US Department of Energy and
National Science Foundation, Istituto Nazionale di Fisica Nucleare
(Italy) and KEK (Japan).
A. Palano is with Universit`
a di Bari, Dipartimento di Fisica and
INFN, I-70126 Bari, Italy
F. Anulli, R. Baldini, A. Calcaterra, R. de Sangro, D. Falciai,
G. Finocchiaro, P. Patteri, I. Peruzzi, M. Piccolo, Y. Xie and A. Zallo
are with Laboratori Nazionali di Frascati dellINFN, I-00044 Frascati,
Italy
S. Bagnasco, A. Buzzo, G. Crosetti, M. Lo Vetere, M. Macri, S. Pas-
saggio, F. Pastore, E. Robutti, A. Santroni, S. Tosi are with Univer-
sit`
a di Genova, Dipartimento di Fisica and INFN, I-16146 Genova,
Italy
C. Cartaro, N. Cavallo, G. De Nardo, F. Fabozzi, C. Gatto,
L. Lista, P. Paolucci, D. Piccolo and C. Sciacca are with Universit`
a
di Napoli Federico II, Dipartimento di Scienze Fisiche and INFN,
I-80126 Napoli, Italy
M. Carpinelli, F. Forti, N. Neri, and E. Paoloni are with Universit`
a
di Pisa, Scuola Normale Superiore, and INFN, I-56010 Pisa, Italy
F. Ferrarotto, F. Ferroni, S. Morganti, G. Piredda, are with Univer-
sit`
a di Roma La Sapienza, Dipartimento di Fisica and INFN, I-00185
Roma, Italy
R. Bionta, V. Brigljevic, D.J. Lange and D.M. Wright are with
Lawrence Livermore National Laboratory, Livermore, CA 94550,
USA
C. Lu is with Princeton University, Princeton, NJ 08544, USA
R. Messner, A. Roodman, A. Soha, J. Vavra and W.J. Wisniewski
are with Stanford Linear Accelerator Center, Stanford, CA 94309,
USA
J.E. Brau, R. Frey, E. Grauges, M. Iwasaki, N.B. Sinev and
D. Strom are with University of Oregon, Eugene, OR 97403, USA
E-mail: strom@physics.uoregon.edu
H.R. Band, A. Eichenbaum and J.R. Johnson are with University
of Wisconsin, Madison, WI 53706, USA
T.B. Moore and H. Neal are with Yale University, New Haven, CT
06511, USA
I. Introduction
The BaBar instrumented ux return (IFR) contains over
2000 m
2
of resistive plate chambers (RPCs) to provide
muon and neutral hadron detection.
The system, com-
posed of 806 modules, is described in detail elsewhere [1]
and is depicted in Figure 1. Both planar and cylindrical
chambers are deployed.
The planar RPC structure is illustrated in Figure 2.
These chambers are constructed of bakelite sheets sepa-
rated by rows of 2 mm thick polycarbonate spacers with
cylindrical symmetry and berglass frame. The inner sur-
faces of the bakelite were treated with linseed oil for surface
smoothing and possible UV absorption [2]. In the BaBar
RPC production at General Tecnica [3], the inside of each
chamber was lled with a 70% oil, 30% pentane mixture
which was then slowly drained. In the original BaBar pro-
duction, this procedure was repeated three times.
The
bakelite was selected to have bulk resistivity of 10
11
to
10
12
cm. The external surfaces of the bakelite are coated
with graphite with surface resistivity of 100k/ . The
RPCs operated in streamer mode, at voltages below 8kV.
The gas in the chambers is 4.5% isobutane, with the bal-
ance a mixture between Ar and Freon R134a (C
2
H
2
F
4
).
The fraction of Freon R134a was adjusted to optimize per-
formance and varied from 45% to 35%. Signals are read
out capacitively, on both sides of the gap, by external elec-
trodes made of aluminum strips on a mylar substrate.
The iron is segmented into 18 plates, giving a total thick-
ness of 65 cm in the barrel and 60 cm in the end caps. A
novel feature of the BaBar detector is the graded segmen-
tation of the iron, which varies from 2 to 10 cm, increasing
with the radial distance from the interaction region. This
segmentation is the result of detailed Monte Carlo stud-
ies which have shown that muon identication at low mo-
mentum and K
0
L
detection improve, for a given amount of
absorber, as the thickness of the iron plates decreases. An-
other important feature of the BaBar detector is the use of
single gap RPC modules. For a xed number of chambers,
the use of single gap chambers optimizes the detection ef-
ciency for K
0
L
and low momentum muons. However, this
optimization causes the detection eciency for high mo- 101
mentum muons to become very sensitive to the eciency
of the outermost layers of RPCs.
Barrel
342 RPC
Modules
432 RPC
Modules
End Doors
19 Layers
18 Layers
BW
FW
3200
3200
920
1250
1940
4-2001
8583A3
Fig. 1. Overview of the IFR system.
In the following the problems seen in the original BaBar
system (section II) are described. Dedicated studies using
test stands (section III) and studies of the detector ma-
terials (section IV) are then discussed. Finally our reme-
diation eorts (section V) and plans for endcap chamber
replacement and upgrade (section VI) are reviewed.
II. Initial Performance
Initial performance of the system was good.
All pla-
nar RPCs were tested shortly after construction and before
shipment from Italy to SLAC, and again prior to installa-
tion at SLAC. Plateau curves were good, eciencies were
high, and dark currents were low. All chambers had dark
currents below 9 µA/m
2
at the voltage producing 90% ef-
ciency. At operating voltage, almost all chamber ecien-
cies were above 96%.
The history of the chamber eciency versus time is
shown in Fig. 3. During initial operation in the summer
of 1999, the system reached unanticipated temperatures of
30 C external, and even higher within the iron gaps con-
taining the RPCs. Currents quickly rose and exceeded the
capacity of the high voltage system, requiring many cham-
bers to be disconnected between July and October of 1999.
A cooling system was installed to restore temperature con-
trol. The cooling system reduced the dark currents so that
Aluminum
X Strips
Insulator
2 mm
Graphite
Insulator
Spacers
Y Strips
Aluminum
H
.
V
.
Foam
Bakelite
Bakelite
Gas
Foam
Graphite
2 mm
2 mm
8-2000
8564A4
Fig. 2. Cross section of a planar RPC.
Average RPC Efficiency
0.2
0.4
0.6
0.8
1
0
100
200
300
400
500
600
700
800
(a) Barrel
1999
June
Jan.
2000
July
Jan.
2001
July
All RPCs
RPCs with eff 10
%
0.2
0.4
0.6
0.8
1
0
100
200
300
400
500
600
700
800
(b) Forward Endcap
1999
June
Jan.
2000
July
Jan.
2001
July
0.2
0.4
0.6
0.8
1
0
100
200
300
400
500
600
700
800
(c) Backward Endcap
1999
June
Jan.
2000
July
Jan.
2001
July
Fig. 3.
Eciency versus time for (a) Barrel chambers, (b) Forward
Endcap chambers and (c) Backward Endcap chambers. The open
triangles show the eciency for chambers with eciency of at least
10% , the closed circles the eciency for all chambers. The increases
in January and August of 2000 are due to changes in the gas. The
variations in the eciency of the Backward Endcap chambers in 2001
are due to interventions designed to test remediation strategies.
almost all of the chambers could be operated again, how-
ever, the dark currents at 20 C were now much higher.
In addition, the average chamber eciency had fallen to
80%.
Starting in 2000, the amount of Freon 134a was reduced
from 45% to 40% and nally to 35% in order to
increase the eciency. This allowed a higher eciency to
be obtained at lower operating voltages as can be seen from
the increases in eciencies in January and August of 2000.
Despite the changes to the gas, the overall eciencies con-
tinued to decline linearly as can be seen from the gure.
This behavior contrasts with the experience of the L3 col-
laboration with similar chambers [4]. After ve years of
operation the eciency of the L3 double gap chambers re-
mained above 96%.
Detailed eciency maps of the individual BaBar cham-
bers showed that much of the eciency loss was primar