The ANTARES optical beacon system

Full text

Page 1

arXiv:astro-ph/0703355v1 14 Mar 2007
The ANTARES Optical Beacon System
M. Agerone, J.A. Aguilarj,∗, A. Albertu, F. Amelix,
M. Anghinolfii, G. Antong, S. Anvary, F. Ardellier-Desagesy,
E. Aslanidese, J-J. Auberte, R. Auerg, E. Barbaritob,
S. Basar, M. Battaglierii, Y. Becherinic,1, J. Beltramelliy,
V. Bertine, A. Bigiw, M. Billaulte, R. Blaesu, N. de Bottony,
M.C. Bouwhuisv, S.M. Bradburyt, R. Bruijnv,ab, J. Brunnere,
G.F. Burgiof, J. Bustoe, F. Cafagnab, L. Caillate, A. Calzase,
A. Caponex, L. Caponettof, E. Carmonaj, J. Carre,
S.L. Cartwrightz, D. Castelu, E. Castorinaw, V. Cavasinniw,
S. Cecchinic,m, A. Ceresb, P. Charvish, P. Chauchotk,
T. Chiarusix, M. Circellab, C. Colnardv, C. Comp` erek,
R. Conigliones, N. Cottiniw,1, P. Coylee, S. Cuneoi,
A-S. Cussatlegrasd, G. Damyk, R. van Dantzigv, G. De Bonisx,
C. De Marzob,2, R. De Vitai, I. Dekeyserd, E. Delagnesy,
D. Denansy, A. Deschampsh, J-J. Destellee, B. Dinkespielere,
C. Distefanos, C. Donzaudy,3, J-F. Drogouℓ, F. Druilloley,
D. Durandy, J-P. Ernenweinu, S. Escoffiere, E. Falchiniw,
S. Favarde, F. Fehrg, F. Feinsteine, S. Ferryn, C. Fiorellob,
V. Flaminiow, K. Fratinii, J-L. Fudad, S. Galeottiw,
J-M. Gallonen, G. Giacomellic, N. Girardu, C. Gojake,
Ph. Gorety, K. Grafg, G. Hallewelle, M.N. Harakehq,
B. Hartmanng, A. Heijboerv,ab, E. Heinev, Y. Helloh,
J.J. Hern´ andez-Reyj, J. H¨ oßlg, C. Hoffmann, J. Hogenbirkv,
J.R. Hubbardy, M. Jaquete, M. Jaspersv,ab, M. de Jongv,
F. Jouvenoty,4, N. Kalantar-Nayestanakiq, A. Kappesg,
T. Kargg, U. Katzg, P. Kellere, E. Kokv, H. Kokv,
P. Kooijmanv,aa, C. Kopperg, E.V. Korolkovaz, A. Kouchnera,
W. Kretschmerg, A. Kruijerv, S. Kuchg, V.A. Kudryavstevz,
P. Lagiere, R. Lahmanng, G. Lamannae, P. Lamarey,
G. Lambarde, J-C. Languillaty, H. Laschinskyg, J. Lavallee,
Y. Le Guenk, H. Le Provosty, A. Le Van Suue, D. Lef` evred,
T. Legoue, G. Lelaizante, G. Limv,ab, D. Lo Prestif,
Preprint submitted to Elsevier 5 February 2008

Page 2

H. Loehnerq, S. Loucatosy, F. Louisy, F. Lucarellix,
V. Lyashukp, M. Marcelinr, A. Margiottac, R. Masullox,
F. Maz´ eask, A. Mazurer, J.E. McMillanz, R. Megnab,
M. Melissase, E. Mignecos, A. Milovanovict, M. Mongellib,
T. Montarulib,5, M. Morgantiw, L. Moscosoy,a, M. Musumecis,
M. Naumann-Godog, C. Naumanng, V. Niesse, T. Noblee,
C. Olivetton, R. Ostaschg, N. Palanque-Delabrouilley,
P. Payree, H. Peekv, A. Perezj, C. Pettaf, P. Piattellis,
R. Pilleth, J-P. Pineaun, J. Poinsignony, V. Popao,
T. Pradiern, C. Raccan, N. Randazzof, J. van Randwijkv,
D. Realj, B. van Rensv, F. R´ ethor´ ee, P. Rewiersmav,2,
G. Riccobenes, V. Rigaudℓ, M. Ripanii, V. Rocaj, C. Rodaw,
J.F. Rolink, H.J. Roset, A. Rostovtsevp, J. Rouxe, M. Ruppib,
G.V. Russof, G. Rusydiq, F. Salesaj, K. Salomong,
P. Sapienzas, F. Schmittg, J-P. Schullery, R. Shanidzeg,
I. Sokalskib, T. Sponag, M. Spurioc, G. van der Steenhovenv,
T. Stolarczyky, K. Streebg, L. Sulake, M. Taiutii,
C. Tamburinid, C. Taoe, G. Terreniw, L.F. Thompsonz,
F. Urbanoj, P. Valdyℓ, V. Valentex, B. Vallagey, G. Vaudainej,
G. Venekampv, B. Verlaatv, P. Verniny,
G. de Vries-Uiterweerdv,aa, R. van Wijkv, G. Wijnkerv,
P. de Witt Hubertsv, G. Wobbeg, E. de Wolfv,ab, A-F. Yaod,
D. Zaborovp, H. Zacconey, J.D. Zornozaj, J. Z´ u˜ nigaj
2

Page 3

aAPC – AstroParticule et Cosmologie, 10, rue Alice Domon et L´ eonie Duquet
75205 Paris Cedex 13, France
bDipartimento Interateneo di Fisica e Sezione INFN, Via E. Orabona 4, 70126
Bari, Italy
cDipartimento di Fisica dell’Universit` a e Sezione INFN, Viale Berti Pichat 6/2,
40127 Bologna, Italy
dCOM – Centre d’Oc´ eanologie de Marseille, CNRS/INSU et Universit´ e de la
M´ editerran´ ee, 163 Avenue de Luminy, Case 901, 13288 Marseille Cedex 9, France
eCPPM – Centre de Physique des Particules de Marseille, CNRS/IN2P3 et
Universit´ e de la M´ editerran´ ee, 163 Avenue de Luminy, Case 902, 13288 Marseille
Cedex 9, France
fDipartimento di Fisica ed Astronomia dell’Universit` a e Sezione INFN, Viale
Andrea Doria 6, 95125 Catania, Italy
gFriedrich-Alexander-Universit¨ at Erlangen-N¨ urnberg, Physikalisches Institut,
Erwin-Rommel-Str. 1, D-91058 Erlangen, Germany
hG´ eoSciences Azur, CNRS/INSU, IRD, Universit´ e de Nice Sophia-Antipolis,
Universit´ e Pierre et Marie Curie – Observatoire Oc´ eanologique de Villefranche,
BP48, 2 quai de la Darse, 06235 Villefranche-sur-Mer Cedex, France
iDipartimento di Fisica dell’Universit` a e Sezione INFN, Via Dodecaneso 33,
16146 Genova, Italy
jIFIC – Instituto de F´ ısica Corpuscular, Edificios de Investigaci´ on de Paterna,
CSIC – Universitat de Val` encia, Apdo. de Correos 22085, 46071 Valencia, Spain
kIFREMER – Centre de Brest, BP 70, 29280 Plouzan´ e, France
ℓIFREMER – Centre de Toulon/La Seyne Sur Mer, Port Br´ egaillon, Chemin
Jean-Marie Fritz, 83500, La Seyne sur Mer, France
mINAF-IASF, via P. Gobetti 101, 40129 Bologna, Italy
nIPHC – Institut Pluridisciplinaire Hubert Curien, Universit´ e Louis Pasteur
(Strasbourg 1) et IN2P3/CNRS, 23 rue du Loess, BP 28, 67037 Strasbourg Cedex
2, France
oInstitute for Space Sciences, R-77125 Bucharest - M˘ agurele, Romania.
pITEP – Institute for Theoretical and Experimental Physics,
B. Cheremushkinskaya 25, 117259 Moscow, Russia
qKernfysisch Versneller Instituut (KVI), University of Groningen, Zernikelaan 25,
9747 AA Groningen, The Netherlands
rLAM – Laboratoire d’Astrophysique de Marseille, CNRS/INSU et Universit´ e de
Provence, Traverse du Siphon – Les Trois Lucs, BP 8, 13012 Marseille Cedex 12,
France
sINFN – Labaratori Nazionali del Sud (LNS), Via S. Sofia 44, 95123 Catania,
Italy
tSchool of Physics & Astronomy, University of Leeds LS2 9JT, UK
uGRPHE – Groupe de Recherche en Physique des Hautes Energies, Universit´ e de
Haute Alsace, 61 Rue Albert Camus, 68093 Mulhouse Cedex, France
vNationaal Instituut voor Kernfysica en Hoge-Energiefysica (NIKHEF), Kruislaan
409, 1098 SJ Amsterdam, The Netherlands
wDipartimento di Fisica dell’Universit` a e Sezione INFN, Largo B. Pontecorvo 3,
56127 Pisa, Italy
3

Page 4

xDipartimento di Fisica dell’Universit` a ”La Sapienza” e Sezione INFN, P.le Aldo
Moro 2, 00185 Roma, Italy
yDSM/Dapnia – Direction des Sciences de la Mati` ere, laboratoire de recherche sur
les lois fondamentales de l’Univers, CEA Saclay, 91191 Gif-sur-Yvette Cedex,
France
zDept. of Physics and Astronomy, University of Sheffield, Sheffield S3 7RH, UK
aaUniversiteit Utrecht, Faculteit Betawetenschappen, Princetonplein 5, 3584 CC
Utrecht, The Netherlands
abUniversiteit van Amsterdam, Instituut voor Hoge-Energiefysica, Kruislaan 409,
1098 SJ Amsterdam, The Netherlands
Abstract
ANTARES is a neutrino telescope being deployed in the Mediterranean Sea. It
consists of a three dimensional array of photomultiplier tubes that can detect the
Cherenkov light induced by charged particles produced in the interactions of neu-
trinos with the surrounding medium. High angular resolution can be achieved, in
particular when a muon is produced, provided that the Cherenkov photons are
detected with sufficient timing precision. Considerations of the intrinsic time un-
certainties stemming from the transit time spread in the photomultiplier tubes and
the mechanism of transmission of light in sea water lead to the conclusion that a
relative time accuracy of the order of 0.5 ns is desirable. Accordingly, different time
calibration systems have been developed for the ANTARES telescope. In this arti-
cle, a system based on Optical Beacons, a set of external and well-controlled pulsed
light sources located throughout the detector, is described. This calibration system
takes into account the optical properties of sea water, which is used as the detection
volume of the ANTARES telescope. The design, tests, construction and first results
of the two types of beacons, LED and laser-based, are presented.
Key words: neutrino telescope, time calibration, optical beacon
PACS: 95.55.Vj, 95.85.Ry
∗Corresponding author
Email address: J.A.Aguilar@ific.uv.es (J.A. Aguilar).
1Now at: y
2Deceased.
3Also at: Orsay – Universit´ e Paris-Sud, CNRS-IN2P3, Institut de Physique
Nucl´ eaire (UMR 8608) ORSAY, F-91406, France
4Now at: University of Liverpool, Dept. of Physics, UK
5On leave at University of Wisconsin – Madison, 53706, WI, USA
4

Page 5

1Introduction
The ANTARES telescope is an underwater neutrino detector being deployed in
the Mediterranean Sea at a depth of 2500 m offshore from Toulon (France) [1].
This detector, which is a first step toward a km3-scale undersea neutrino tele-
scope, will consist of twelve lines with a sensitive area for high energy muons
of more than 0.05 km2for Eµ> 100 TeV. The construction of the ANTARES
neutrino telescope started with the installation of the first lines in 2006, and
the detector is scheduled to be completed by the end of 2007. In addition,
a special instrumentation line, the MILOM [2], is in operation since Spring
2005. The detection principle and the main detector components are briefly
described in section 2. The precision required in the time determination and
the different time calibration systems are reviewed in section 3. The concept of
an Optical Beacon system is briefly introduced and the actual solution adopted
by ANTARES is reviewed in section 4. Detailed descriptions of the LED and
Laser Beacon calibration systems are given in section 5 and 6, respectively.
Some results from the first data taken with the lines in operation in 2006 are
given in section 7. Finally, section 8 presents the summary and conclusions.
2The ANTARES Neutrino Telescope
The ANTARES neutrino telescope uses sea water as the detection medium to
look for extra-terrestrial neutrinos. Most of these neutrinos cross right through
the Earth without interacting. A small fraction of the incoming neutrino flux,
however, interacts with the nucleons that make up the matter surrounding
the detector. In a charged current interaction a high energy muon neutrino
produces a muon which induces Cherenkov light when crossing a suitable
optical medium such as ice or water. Other signatures can also be detected.
In order to detect and reconstruct the wavefront of the Cherenkov light, AN-
TARES is equipped with 900 Optical Modules (OMs). The OM, the basic
optical unit of ANTARES, consists of a photomultiplier tube (PMT) housed
in a water-pressure resistant glass sphere [3]. An exhaustive study of PMTs
was carried out during the R&D phase which led to the selection of the 14-
stage, 10” Hamamatsu R7081-20 model [4]. Together with the PMT there is
an internal LED for calibration purposes inside the OM. Each group of three
OMs constitutes a storey. All the electronics for one storey are housed in a
pressure resistant titanium container making up the so-called Local Control
Module (LCM). Every OM is read out by an electronics board housed in the
LCM carrying a pair of Analogue Ring Samplers (ARS), the ASIC chip used
for signal processing and digitisation [5]. The ARS provides the time and am-
plitude of the signal, both of which are essential to reconstruct the muon track
5

Page 6

direction and estimate its energy.
3The ANTARES time calibration systems
ANTARES is expected to achieve very good angular resolution (< 0.3◦for
muon events above 10 TeV). This pointing accuracy is closely related to the
precision in the determination of the arrival time of the Cherenkov photons at
the PMTs. The relative time resolution between OMs is, therefore, of utmost
importance. It is limited by the transit time spread (TTS) of the signal in
the PMTs (σ ∼ 1.3 ns) and by the scattering and chromatic dispersion of
light in sea water (σ ∼ 1.5 ns for a light propagation of 40 m) [6,7]. The
electronics of the ANTARES detector is designed in order to contribute less
than 0.5 ns to the overall time resolution. Therefore, the time calibration
should aim at a precision below the nanosecond level. To this end, several
complementary time calibration systems are implemented in the ANTARES
detector in order to measure and monitor the relative times between different
components of the detector due to, e.g. cable lengths and electronics delays.
These time calibrations are performed by the following systems:
(1) The internal clock calibration system. A very precise time reference
clock distribution system has been implemented in the ANTARES detec-
tor. It consists of a 20 MHz clock generator on shore, a clock distribution
system and a clock signal transceiver board placed in each LCM. A com-
mon clock signal is provided to the ARSs. Synchronised data commands
can be superimposed on the clock signal, in particular start and stop com-
mands, which together with a high precision Time to Digital Converter
(TDC) make up the essential components of the system. This system
also includes an echo-based time calibration whereby each LCM clock
electronics board is able to send back a return signal through the same
optical path as the outgoing clock signals. This system enables the time
offsets between all LCM clock boards to be measured by recording the
propagation delays of the return signals of each storey with respect to
the original clock signal emission time. Measurements in real conditions
show a resolution of ∼ 0.1 ns, well within the specifications. The system
also includes the synchronisation with respect to Universal Time, by as-
signing the GPS timestamp to the data, with a precision of about 100 µs,
much better than the required precision of ∼ 1 ms. The clock signals are
distributed across all detector components from the shore up to the clock
boards. The remaining path between these boards and the PMT photo-
cathodes however requires a different timing calibration mechanism.
(2) The internal Optical Module LEDs. Inside each Optical Module there
is a blue LED attached to the back of the PMT capable of illuminating the
photocathode. The LED is an HLMP-CB15 from Agilent whose light in-
6

Page 7

tensity is peaked at around 470 nm with a FWHM of 15 nm. These LEDs
are used to measure the relative variation of the PMT transit time and
dedicated runs of this LED calibration system are customarily taken [2].
This system is used to calibrate the path travelled by the signal starting
at the PMT photocathode up to the read-out electronics. The effect of
the transmission of the light in water is, however, not addressed by this
calibration method.
(3) The Optical Beacons. This system allows the relative time calibra-
tion of different OMs to be determined by means of independent and
well-controlled pulsed light sources. This system also makes possible to
monitor the influence of the water on the light propagation. These Opti-
cal Beacons are the subject of this paper and will be described in detail
in the following sections.
(4) Several thousands of down-going muon tracks will be detected per day.
The hit time residuals of the reconstructed muon tracks can be used to
monitor the time offsets of the Optical Modules. This methodology will
enable an overall space-time alignment and calibration cross-checks.
Prior to the deployment of the lines, all line elements are verified as functioning
correctly in a dedicated dark setup where a time calibration is carried out
after the integration of each sector of the line (a sector is one fifth of a line).
An optical signal is sent to each OM of every storey. The signal is provided
by a Nd-YAG solid state laser that emits intense, short duration light pulses.
The light pulse is attenuated before being sent to the OMs. The light is guided
through an optical fibre to a 1-to-16 optical splitter. Each of the outgoing fibres
is connected to one of the 15 OMs of the sector. The 16thsignal is sent to a
control module and is used as a time reference. The resulting information from
timing calibration in the dark setup is used as the reference for the validation
of the in situ timing calibrations. This system is also used to determine the
time calibration of the Optical Beacons.
The time calibration depends on the actual location of the OMs which is
affected by the slow movements of the lines due to underwater currents. An
acoustic positioning system together with a set of compasses and tiltmeters
located along the line, provides the OM position with an accuracy of 10-
20 cm which, in addition to the time calibration, is sufficient for the muon
track reconstruction [2].
4 The ANTARES Optical Beacon system
The Optical Beacon system consists of a series of pulsed light sources dis-
tributed throughout the detector. An LED Beacon is composed of several
LEDs, pulsed by dedicated electronic circuits. Those beacons are located, more
7

Page 8

or less, uniformly along every detector line so that their light can illuminate all
storeys on the neighbouring lines. The Laser Beacons use a solid state pulsed
laser whose light is spread out by a diffuser. Laser Beacons are located at
the bottom of a few lines in the so-called Bottom String Socket. The Laser
Beacons illuminate mainly the bottom part of the lines and are located in a
stationary position. The system of Optical Beacons provides a number of well
controlled, pulsed light sources that act as a reference for time calibration of
the detector. The system is able to closely monitor all the detector components
and the sea water. It allows the monitoring of the absorption and scattering
lengths of the sea water.
5 The LED Beacons
An LED Beacon contains 36 individual LEDs arranged in groups of six, on
six vertical boards (faces) which are placed side by side forming an hexagonal
cylinder (figure 1). On each face, one LED points upwards (top LED), and the
other five LEDs point radially outwards. One of the LEDs that points hori-
zontally is located in the middle of the face (central LED) and the remaining
four surround it.
Fig. 1. An LED Beacon as viewed from sideways (left) and from the top (right). Six
faces each containing six LEDs are arranged on an hexagonal cylinder. The internal
photomultiplier tube is mounted at the centre of the lightguide.
The faces are mechanically fixed to a hollow nylon structure which internally
houses a small Hamamatsu H6780-03 photomultiplier tube. This PMT, with
8

Page 9

a photocathode of 8 mm diameter, has a risetime of 0.8 ns and a transit
time of 5.4 ns and is used to provide the precise time of emission of the
light flash independently of the triggering signal. A flat acrylic disc that acts
as a lightguide is fixed to the upper part of the nylon mounting to increase
the collection of light. Following Fields and Janowski [9] a conical depression
was machined in the centre of the light collecting disc, to direct light into
the photomultiplier tube. The edges of the disc were also bevelled at 45◦to
improve light collection from the horizontal LEDs.
The lower part of the LED Beacon houses the electronic boards that provide
the required operating voltages and enable the actual LED flashing according
to externally supplied slow control commands. Each of the six faces can be
flashed independently or in combination of different faces. Within a face the
top, central and the group of four LEDs can be triggered independently or in
combination. This layout allows a distribution close to uniform in the azimuth
angle when all the LEDs are flashing. The top LEDs allow the calibration of
the OMs in those storeys on the same line above the beacon. The amount
of light can be further controlled changing the number of LEDs flashing at
a given time while the intensity of the LEDs can also be varied as explained
in the following subsection. The possibility of flashing certain faces enables
the monitoring of the uncertainty in the time calibration arising from the
non-uniformity and anisotropy of the LED Beacons.
5.1 The pulser system
The pulser circuit is based on an original design from Kapustinsky et al. [10]
that has been modified for ANTARES to include in particular a variable ca-
pacitor that enables the synchronisation of the pulses produced by several
different circuits (see figure 2). The trigger is provided by a 1.5 V negative
square pulse of a duration of around 150 ns superimposed on a negative DC
bias that can be varied from 0 to 24 V. The DC component charges the capac-
itor and the rising edge of the differentiated 1.5 V pulse switches on the pair
of transistors, triggering the fast discharge of the 100 pF capacitor through
the low impedance path that includes the LED. The parallel inductor de-
velops charge in opposition to the discharging capacitor further reducing its
time constant. The level of the DC voltage determines the amount of current
through the LED and thus the intensity of the emitted pulse. The layout of
the printed circuit board has been designed so as to enable the inclusion of
six pulsers on the same face without interference of the distributed triggering
signals in the nearby pulser circuits whilst minimizing the difference in the
times of arrival of the trigger signal to the different pulsers. Synchronisation
of the signals from different LEDs on the same or on different faces is possible
by adjusting the variable capacitor in each pulser, this procedure is explained
9

Page 10

in subsection 5.3. When deployed underwater the LED Beacons are operated
with a typical trigger frequency of a few Hz. This frequency can be increased
to rates up to 1 kHz.
Fig. 2. Diagram of the LED pulser circuit. The variable capacitator (see text) is
indicated by label 3C4.
5.2 The LEDs
Different types of LEDs were tested by the ANTARES collaboration in terms
of amplitude and risetime of the emitted light pulses. The selected LED was
the Agilent HLMP-CB15-RSC00 model6. This LED has a peak wavelength
of 472 nm with a spectral half-width of 35 nm according to the specification
sheet. The risetime of the LED pulses has been measured and found to be
between 1.9 and 2.2 ns. The LEDs were classified in batches according to
their risetime and the fastest (1.9-2.0 ns) are used for the top LED locations
where the uncertainty in the calibration is dominated by the rise-time and not
by the light propagation effects.
To increase the angular occupancy of the light emitted by the LEDs, which
was originally restricted to 15◦, the caps of the LEDs were machined off.
The angular distribution of the emitted light for several depths of cut was
measured. The cut lengths tested ranged from 1.5 to 3.5 mm. A cut at 3 mm
was selected, which provides an emission flat within ±10% for angles up to
35◦and within a factor two up to 55◦.
Figure 3 (left) shows the light amplitude as function of the azimuth angle for
different cut depths. The right plot in figure 3 shows the time distribution of
a single LED pulse.
6Agilent Technologies, Inc. Headquarters 395 Page Mill Rd. Palo Alto, CA 94306
United States.
10

Page 11

-60-40-200 20
angle [degree]
4060
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
1.5 mm
2.5 mm
2.0 mm
3.0 mm
3.5 mm
Light yield (a.u.)
time [ns]
-6-4-20246810 12
amplitude [V]
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Fig. 3. Left: Light amplitude as a function of the azimuth angle for different cut
depths. Right: Timing profile of a single LED pulse.
As mentioned in the previous subsection, the amount of energy per pulse
emitted by the LED can be varied by changing the DC voltage from 0 to
24 V. Below 8 V the amount of light emitted is almost negligible. For a DC
level of 24 V the energy per pulse emitted is at least 150 pJ, which corresponds
to the emission of approximately 4 · 108photons.
5.3Assembly, synchronisation, testing and integration
Once an LED Beacon is assembled, its 36 LEDs have to be synchronised by
tuning the variable capacitor in each of its pulser circuits. This operation
takes place in three steps. First the achievable time range of light emission
is measured for each LED by adjusting the variable capacitor to its upper
and lower limits. Then, taking advantage of the overlap of these time ranges,
a common reference time for all LEDs is chosen. Finally, the capacitors are
tuned again to reach this reference time. In the left-hand plot of figure 4 the
different ranges for each LED pulser for a typical LED Beacon are shown.
The full (red) line indicates where the synchronisation common reference time
was set and the squares (blue) are the final measured times. The right-hand
plot of figure 4 shows the distribution of the final emission times. The typical
standard deviation of the emission times is in general a few tens of picoseconds.
Following this synchronisation procedure, the beacon undergoes a series of
thermal cycles in a climate chamber lasting 48 hours in order to guarantee
its stability. The emission times are then remeasured and in exceptional cases
some pulsers are re-synchronised.
After synchronisation, the LED Beacon is introduced into the pressure resis-
11

Page 12

Relative synchronization time (ns)
1.31.321.341.361.38 1.4
entries
0
1
2
3
4
5
6
7
8
Mean: 1.35 ns
Sigma: 0.01 ns
Fig. 4. Left: Emission time ranges of the 36 LEDs of a typical LED Beacon. The
horizontal straight line indicates the desired common emission time (reference time),
the small squares show the final measured emission times. Right: Distribution of the
final emission times.
tant glass vessel that will house it in the sea. This is a cylindrical borosilicate
glass container commercially available7. It consists of a cylinder and two end-
caps (figure 5 left), one of them detachable. All the parts are in transparent
glass except for the titanium flanges that hold the two parts together. The
overall dimensions of the cylinder plus endcaps are 210 mm outer diameter
and 443 mm in length. The endcap is supplied with a 22 mm diameter pre-
drilled hole equipped with a penetrator on the outside of the cylinder and
connecting cables on the inside. The LED Beacon is mechanically attached to
the detector lines by a collar mounted on the Optical Module Frame (OMF).
It is held vertically at a specific location above the triplet of OMs (see figure 5
right) and fixed to the structure combining the 6-fold symmetry of the beacon
and the 3-fold symmetry of the OMF so as to minimise shadowing.
As already mentioned, the lines undergo a calibration procedure at the inte-
gration sites using a common laser source. The PMT of the LED Beacon is
calibrated simultaneously with the PMTs in the OMs using a dedicated fibre
from this common source. This calibration facilitates the measurement of their
relative shift in the arrival times.
7Nautilus Marine Service GmbH, Blumenthalstrasse 15 D-28209 Bremen Germany.
12

Page 13

Fig. 5. Left: Example of a borosilicate glass container that houses the LED Beacons.
Right: LED Beacon mounted in the Optical Module Frame.
5.4Configuration, control and monitoring
The LED Beacon motherboard controls the operation of the beacon at the
most basic level. Its main component is the UNIV1 card, a control board used
throughout ANTARES which is based on a 17C756 PIC controller. Commu-
nication with the motherboard takes place on an RS485 serial link protocol
MODBUS.
The motherboard main functions are 1) to set the DC level supplied to the
LEDs (from the +48 V input voltage via DC/DC converter) and thus the
intensity of the light pulses emitted, 2) to select the faces or group of faces
that will be flashed, 3) to select any of the three groups of LEDs or the
combination thereof that will flash, 4) to set the PMT gain control voltage, and
5) to monitor the voltage supplied to the LEDs and the ambient temperature
as measured by a sensor.
The configuration of all the LED Beacons in the detector is performed through
the general RunControl program [11]. Different configurations corresponding
to the set of possible faces and groups of LEDs that are requested to flash
and the required PMT and LED voltages, are stored in a database. In the
configuration stage at the start of a calibration run, a given configuration (a
run set-up) can then be selected and downloaded. In addition, a graphical user
interface written in Java allows the expert user to communicate directly with
the beacons and request their status or change their configuration during the
run by generating the necessary Slow Control commands. This latter method is
very rarely used during normal data taking, but is employed in the laboratory
for test or debugging purposes.
13

Page 14

6The Laser Beacons
The Laser Beacons emit high intensity, short duration pulses of light and
will be located at the bottom of a few lines, attached to their Bottom String
Socket (BSS), i.e. the mechanical structure that anchors the line to the sea
bed. At present, one Laser Beacon has been installed in the so-called MILOM
line [2] (see next section). Figure 6 shows a general view of the Laser Beacon
and its components. The Laser Beacon points upwards so that the emitted
light can reach nearby lines. Since the intensity and time profile of the light
pulse are defined by the intrinsic properties of the laser, an extensive study of
different laser types was made. The selected model is described in the following
subsection.
Fig. 6. A Laser Beacon dismantled: On the left the inner mechanics holding the
laser head and its associated electronics; on the right the pressure-resistant titanium
container that houses the equipment. On the top endplate (right part of the picture)
of the container the quartz cylinder that prevents sedimentation effects can be seen.
6.1 The laser
The main component of the Laser Beacon is a diode pumped Q-switched Nd-
YAG laser which produces short pulses with a time duration less than 1 ns
(FWHM) and a total energy of ∼1 µJ. The laser model selected is the NG-
10120-1208which emits at 532 nm after frequency doubling of the original
Nd-YAG wavelength of 1064 nm.
The laser is very compact. Its head dimensions in mm are 144×37.4×30. The
laser can be operated in a non-triggered mode at a fixed frequency (around
8Nanolase, presently part of JDS Uniphase Corp., 430 N. McCarthy Blvd. Milpitas,
CA 95035 United States.
14

Page 15

15 kHz) or in a triggered mode with a variable trigger frequency. In the latter
mode, which is the one being used in ANTARES, the laser is triggered when
a TTL signal arrives at the device through a connection in the rear panel of
the power supply. Since the laser is passively Q-switched, the delay between
the trigger signal and the light pulse emission is of the order of microseconds
and the pulse to pulse jitter is of the order of a few hundred nanoseconds. The
actual time of laser emission is obtained thanks to a fast photodiode integrated
into the laser head.
Once the laser shot is produced, the built-in photodiode sends back a signal
which is passed to an ARS chip located in the String Control Module (SCM),
the electronics container similar to the LCM located on the BSS. The current
that feeds the pumping diode is switched off and the system waits for the next
trigger signal.
The power supply delivered with the laser was refurbished in order to comply
with the technical requirements of the experiment and, at the same time, to
accommodate the whole apparatus into a smaller space. The signal from the
photodiode is reshaped electronically to fulfill the constraints imposed by the
front-end electronics of the experiment.
In order to characterise the relevant features of the Nd-YAG laser, a thorough
study of the main laser parameters was made:
(1) The intrinsic jitter of the Q-switching mechanism gives rise to a jitter
in the laser pulse emission time of a few hundred nanoseconds. It was,
therefore, necessary to confirm that the time recorded by the internal
photodiode was sufficiently accurate for our needs. Several fast external
photodiodes (Newport 818-BB-20, Alphalas UPD-200-SP, Hamamatsu
S5973-01) were employed to estimate the accuracy in the emission time
given by the internal photodiode. Figure 7 (top-left) illustrates the dif-
ference in emission time as measured by an external Newport 818-BB-20
photodiode and the internal built-in photodiode. The standard deviation
of the distribution is 50 ps (the position of the peak is immaterial, it
depends on delays that will be determined by calibration).
(2) The pulse shape was measured using a Hamamatsu streak camera. In
figure 7 (top-right) an example of a pulse of the laser as sampled by the
streak camera (5 ps resolution) is shown. The FWHM of the pulse is
determined to be smaller than 0.8 ns. The time shape profile is smooth
and not far from Gaussian. As expected, the timing features of the pulse
did not change with the three trigger frequencies studied, namely 100 Hz,
1.5 kHz and 10 kHz.
(3) Different energy measurements were also performed with a special de-
vice9capable of measuring the energy of each pulse. As can be seen in
9A photodiode head Model PD10 and a laser power meter LaserStar from Ophir,
15