Elsevier

Astroparticle Physics

Volume 72, 1 January 2016, Pages 61-75
Astroparticle Physics

The major upgrade of the MAGIC telescopes, Part I: The hardware improvements and the commissioning of the system

https://doi.org/10.1016/j.astropartphys.2015.04.004Get rights and content

Abstract

The MAGIC telescopes are two Imaging Atmospheric Cherenkov Telescopes (IACTs) located on the Canary island of La Palma. The telescopes are designed to measure Cherenkov light from air showers initiated by gamma rays in the energy regime from around 50 GeV to more than 50 TeV. The two telescopes were built in 2004 and 2009, respectively, with different cameras, triggers and readout systems. In the years 2011–2012 the MAGIC collaboration undertook a major upgrade to make the stereoscopic system uniform, improving its overall performance and easing its maintenance. In particular, the camera, the receivers and the trigger of the first telescope were replaced and the readout of the two telescopes was upgraded. This paper (Part I) describes the details of the upgrade as well as the basic performance parameters of MAGIC such as raw data treatment, linearity in the electronic chain and sources of noise. In Part II, we describe the physics performance of the upgraded system.

Introduction

MAGIC (see Fig. 1) is a stereoscopic system of two Imaging Atmospheric Cherenkov Telescopes (IACTs) located at 2200 m a.s.l. in the observatory of Roque de los Muchachos in La Palma, Canary Islands (Spain). Together with the H.E.S.S. IACTs in Namibia [1] and the VERITAS IACTs in Arizona [2], MAGIC is the most sensitive instrument for high-energy gamma-ray astrophysics in the range between few tens of GeVs and tens of TeVs.

Contrary to optical telescopes, IACTs observe dim (∼100 photons/ m2/ TeV) short (∼ns) flashes produced by extended air showers developing in the atmosphere (see reviews by, e.g., [3], [4]). The light, mostly emitted in the UV and optical wave bands, is produced via Cherenkov radiation from the charged particles of the atmospheric shower, which travel faster than the light in the air. The amount of Cherenkov light and its angular and spatial distribution carry information about the energy and incoming direction of the primary cosmic rays and γ rays, which is reconstructed analyzing the image formed on the focal plane of the IACTs. The images roughly resemble an ellipse, whose brightness, geometrical size, and orientation represent the most basic parameters used in the subsequent data analysis (see [5] for details). The telescopes are self-triggered by multiple (neighbor) pixels above a certain signal threshold. Because the Cherenkov light flashes from air showers are very short, typically few nanoseconds long, the use of extremely fast and sensitive light sensors, typically photomultiplier tubes (PMTs), and fast electronics for the trigger and signal sampling is the key to discriminate the shower light from fluctuations of the night sky background. The amount of Cherenkov photons reaching the pixels is reconstructed from the signal charge in the PMTs, by analyzing the ultra-fast sampled snapshot of the signal pulse. An “extraction” method, that basically sums the ADC counts in a certain time (sliding) window, provides a rough signal charge per channel, which, after a calibration procedure, is converted into the number of photons at the camera plane [6]. A coincidence (stereo) trigger among individual telescopes minimizes spurious events triggered by the night sky background light, triggers by the so-called afterpulsing effect of the PMTs or by single local muons flashing only one telescope. Moreover, in the so-called stereoscopic reconstruction, multiple images of the air shower allow the energy and the incoming direction of the primary γ ray to be more precisely reconstructed.

The two MAGIC telescopes started operation 5 years apart (MAGIC-I in 2004 and MAGIC-II in 2009, respectively), and the second telescope was an “improved clone” of the first one. The main reasons for differences were funding constraints during the building of the first telescope and the technological progress that took place in the years between the design of the two telescopes. The major goal of the telescopes is a lowest possible energy threshold, which is achieved through fine pixelated cameras, fast sampling electronics and a large mirror area. The second goal is a fast repositioning speed in order to catch rapid transient events such as Gamma-Ray Bursts, which is achieved through a light weight (<70 tons) telescope structure made out of reinforced carbon fibre tubes. The structure requires an automatic mirror control (AMC) to maintain the best possible optical point spread function at different zenith angles of observations [7], [8]. The readout and the trigger electronics are located in a dedicated counting house, where the signals transmitted via optical fibers from the cameras are received. A difference in transit time between signals in different channels (mainly due to different high voltages applied to PMTs) is corrected online at trigger level by means of adjustable delay lines to minimize the needed trigger gate and offline for the reconstruction of the signal arrival time and charge. The achieved energy threshold is as low as ∼50 GeV at the trigger level for observations at zenith angles below 25° (see Fig. 6 in [9]). This energy threshold is achieved by means of a digital trigger. Using the so-called sum-trigger, it is possible to reach an even lower energy threshold [10], and a new version of the sum-trigger is currently under commissioning [11]. The repositioning speed is maintained throughout the years to be ∼25 s for a 180° rotation in azimuth.

While the above mentioned concepts made the two MAGIC telescopes look very similar there were few important design differences between MAGIC-I and II before the upgrade described in this paper. Funding permitted to equip the entire MAGIC-II field of view (FoV) homogeneously with small 1 inch PMTs, compared to the mixed 1 and 2 inch pixel configuration of the MAGIC-I camera. The active trigger area, which in all MAGIC cameras is limited to a central area in the FoV, was enlarged from ∼0.9° radius (trigger area of the old MAGIC-I camera) to ∼1.2° radius (in the MAGIC-II camera), still using the same trigger electronics as for the MAGIC-I camera but reducing the size of overlapping sectors (see Section 3). The main motivation for enlarging the sensitive trigger area was to adapt to the stereo approach and increase sensitivity to extended γ-ray sources as well as a more suitable usage of the so-called wobble mode (pointing to a source of interest at some 0.4° off-center, [12]) for a better background estimation.

In detail, the main resulting differences between the two telescopes were the following ones:

  • The camera of the MAGIC-I telescope consisted of 577 PMTs divided in 397 small PMTs, 1 inch diameter each, in the inner part of the camera and 180 large PMTs, 2 inch diameter each, in the outer part. The FoV the camera was 3.5°. The camera of MAGIC-II consists of 1039 PMTs, all 1 inch diameter, and has the same FoV as the first camera.

  • The region of the MAGIC-II camera exploited for the trigger was 1.7 times larger than the one of MAGIC-I.

  • The MAGIC-I readout was based on an optical multiplexer and off-the-shelf Flash Analog to Digital Converters (FADCs) (MUX-FADC, [13]), which was robust and had an excellent performance but was expensive and bulky. The readout of MAGIC-II was based on the DRS2 chip2 (compact and inexpensive but performing quite worse in terms of intrinsic noise, dead time and linearity compared to the MUX-FADC system).

  • The receiver boards of MAGIC-I (see Section 3.3.1), the part of the electronics responsible to convert the optical signals coming from the camera and to generate the level zero trigger signal, lacked programmability. They were also showing high failure rate, mainly due to aging.

In 2011–2012 MAGIC underwent a major upgrade program to improve and to unify the stereoscopic system of the two telescopes. Most importantly, the camera of MAGIC-I was replaced by a new one, the readout of the two telescopes replaced by a more modern system, and the trigger area of the MAGIC-I camera was increased to match the one of MAGIC-II. Table 1 provides a brief summary of the most relevant hardware characteristics of the telescopes before and after the upgrade. This paper (Part I) describes the motivation for the upgrade, its main steps, the commissioning of the system and the low level performance of MAGIC. In Part II [9] we describe the physics performance of the upgraded system.

Section snippets

Motivation for the upgrade

There were three main motivations for the upgrade of the MAGIC system. The first one was the wish to improve the stereoscopic performance of the MAGIC system. Several key parameters were targeted for improvement:

  • The low energy performance. The performance of MAGIC to the lowest accessible energies was limited by the electronic noise in the DRS2 system of the MAGIC-II telescope. With a lower noise system the analysis energy threshold can be lowered, and the performance close to the threshold

Individual parts of the upgrade

In this section we describe the main hardware parts that have been upgraded. The individual hardware items of the upgrade program are shown in Fig. 2 .

Low level performance

Here we shortly describe the basic performance parameters of the MAGIC telescope system after the upgrade.

Commissioning of the system

The key point of the efficient commissioning was to have a dedicated and well experienced team of 5–10 physicists at the site of the experiment for a duration of several months after the installation of the hardware. In the following the main milestones of the commissioning are described.

The online analysis client

A real time data analysis is an important part of the success of an IACT experiment. Most of the extragalactic and several Galactic very-high-energy sources are variable, some of them on time scales down to hours and minutes. A real time analysis of the data taken can provide essential time critical internal triggers to extend observation of flaring sources and alert other multiwavelength partners.

The upgraded system allowed to develop a novel program to fulfill the task of analyzing the data

Conclusions

A major upgrade of the MAGIC telescopes took place in the years 2011–2012. The major items were the installation of the new camera for MAGIC-I, the new trigger in the MAGIC-I telescope, the upgrade of the readout system to DRS4 and programmable receiver boards in both telescopes. The commissioning of the upgraded system successfully finished in October 2012, and the telescopes restarted regular operation. The main goals of the upgrade were an improvement of the sensitivity at low energies,

Acknowledgements

We would like to thank the Instituto de Astrofísica de Canarias for the excellent working conditions at the Observatorio del Roque de los Muchachos in La Palma. The financial support of the German BMBF and MPG, the Italian INFN and INAF, the Swiss National Fund SNF, the ERDF under the Spanish MINECO, and the Japanese JSPS and MEXT is gratefully acknowledged. This work was also supported by the Centro de Excelencia Severo Ochoa SEV-2012–0234, CPAN CSD2007–00042, and MultiDark CSD2009–00064

References (31)

  • F. Aharonian et al.

    Observations of the Crab nebula with HESS

    Astron. Astrophys.

    (2006)
  • J. Holder, V.A. Acciari, E. Aliu, et al., Status of the VERITAS Observatory, in: F.A. Aharonian, W. Hofmann, F. Rieger...
  • J. Hinton

    Ground-based gamma-ray astronomy with Cherenkov telescopes

    New J. Phys.

    (2009)
  • E. Lorenz et al.

    Very-high energy gamma-ray astronomy. A 23-year success story in high-energy astroparticle physics

    Eur. Phys. J. H

    (2012)
  • A.M. Hillas

    Cerenkov light images of EAS produced by primary gamma

    Int. Cosmic Ray Conf.

    (1985)
  • Cited by (296)

    • Photo-Trap: A low-cost and low-noise large-area SiPM-based pixel

      2023, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    • Pointing calibration of LHAASO-WFCTA telescopes using bright stars

      2023, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    • Status and perspectives of vacuum-based photon detectors

      2023, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    • Highlights of the very-high-energy gamma-ray sky as seen by MAGIC

      2023, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    View all citing articles on Scopus
    1

    Deceased.

    View full text