Title: Finalizacin y explotacin de los telescopios MAGIC Completion and Exploitation of the MAGIC Telescope
1Finalización y explotación de los telescopios
MAGICCompletion and Exploitation of the MAGIC
Telescopes
Juan Cortina IFAE Barcelona
FPA2009-07474
2- Scheme of the talk
- MAGIC-I some scientific highlights.
- MAGIC-II construction and current status.
- MAGIC-I upgrade.
- Description of the project.
3MAGIC-I
The MAGIC-I Imaging Atmospheric Cherenkov ?-ray
Telescope was built is operated by a
collaboration of 150 Physicists in
IFAE Barcelona, UCM Madrid, UA Barcelona, U
Barcelona, IEEC, Instituto Astrofísica de
Canarias, Instituto Astrofísico Andalucía MPI
Munich, U. Würzburg, DESY Zeuthen, U. Dortmund,
INFN/U. Padua, INFN Pisa/Siena, U. Udine, INAF
(Italy), ETH Zurich, U. Lodz (Poland), Tuorla
Observatory (Finland), UC Davis (US), Physics
Institute Sofia (Bulgaria), Croatian
Consortium. Location La Palma, Canary Islands,
Spain
4Emphasis on high tech
5MAGIC-I performance
- Energy resolution ?E/E 25 above 100 GeV
- Point-source flux sensitivity in 50 hours 1.6
crab - Lowest energy threshold IACT 25-50 GeV new
trigger allows overlap with satellites (Fermi)
6IFAE _at_ MAGIC-I
- Major contributions to telescope
- Camera RD and construction
- Camera Control and Central Control
- Telescope Calibration System
- Camera Access Tower
- MAGIC Control House
- Telescope commissioning
- Good part of Analysis and MC software
- Telescope characterization
- Physics Analysis
- Common Fund administration
7IFAE _at_ MAGIC-I
- IFAE has also organizational responsabilities
- J.Cortina
- Technical coordinator (2005-06) Operations
coordinator (2008-) - Convener Galactic Group (2006-08)
- Deputy Physics Coordinator (2007-08)
- One of the 7 members of Executive Board
(2005-06, 2008-) - M.Martinez
- Co-spokesperson (1998-2003)
- Spokesperson (2003-2005)
- Chairman of Collaboration Board (2008-)
- One of the 7 members of Executive Board (2008-)
- A. Moralejo
- Software coordinator (2006-)
- One of the 7 members of Executive Board (2006-)
- J. Rico
- Convener Galactic Group (2008-)
8Four years of data
- We are about to complete our 4th year of
observations - About 4000 hours dark time (high efficiency 65)
- About 1300 hours of moon time (impossible by
other telescopes) - We have published 47 papers in the last 3 years,
more than one paper/month. - Among them we have published 3 papers in Science
- First variable microquasar at Very High Energies
(VHE) a quasar at short distance, i.e. easier to
study. - The farthest quasar at VHE impact on density of
Extragalactic Background Light. - First pulsar discovered at VHE most energetic
photons ever emitted by a pulsar.
9MAGIC-I scientific highlights
- I focus on only a few relevant results
- ?-ray binaries LSI 61 303.
- Quasar 3c279 and implications for Extragalactic
Background Light. - Flare of Mrk 501 and Lorentz invariance.
- More highlights will be presented in next talk.
10Binary system LS I 61 303
11VHE radiation from LSI 61
- First discovered by MAGIC Science 312 (2006),
corresponding author N. Sidro (IFAE)
12LSI 61 303 Periodicity
- Astrophys. J. 693 (2009) 303 and PhD thesis of
N. Sidro (IFAE) - Source is periodic in VHE with a period
consistent with orbital period correlation to
relative position of the two stars.
13Attenuation of ?-rays in the Extragalactic
Background Light
AGN
?max 1.24 ?m (E / 1TeV)
14Spectrum of Extragalactic Background Light
- An imprint of the history of the Universe (star
formation, galaxy evolution) - Direct measurements challenging (foregrounds)
- ? Large uncertainties
Redshifted star light
Redshifted dust emission
153C 279 (z 0.536) the most distant VHE ?-ray
source
Science 320 (2008), corresponding author D.
Mazin (IFAE)
- MAGIC detects a clear signal from 3C 279 in
February 2006 - First quasar detected at VHE
- Measurements constrain EBL models under safe
assumptions about the intrinsic spectrum
16EBL constraints
17Tests of speed of light invariance
L
observer
source
- Need fast-varying source (AGN flares...)
18Results before MAGIC
Pulsars, E up to 2 GeV, L about 10 kpc, (Kaaret,
1999)
Mrk 421, E up to 10 TeV, L about 100 Mpc
(Biller et al, 1999)
35 GRBs, E up to MeV, L beyond 7000 Mpc (Ellis
et al, 2005)
19MAGIC observations of Mrk 501 in July 2005
Astrophys. J. 669 (2007)
- z 0.034
- (L?100 Mpc)
- Hint (2.5 ?) of energy-dependent delay of fast
flare
20MAGIC results Phys.Lett B 668 (2008)
Linear
Quadratic
MQG lower limits (95 C.L.)
gt 6 times better than previous limits
gt 4 times better than previous limits
Strong involvement of IFAE (M. Martínez, PhD
theses of J. López, O. Blanch and M. Errando, ApP
19 (2003) 245-252, arXiv0803.2120 -subm. to NIM)
21(obvious) caveat of method
- Source-intrinsic E-dependent delays are possible.
- Limits depend on the assumption of simultaneous
emission at all Es - Many sources at different distances are needed
22Upgrade to stereo MAGIC-II
23MAGIC-II
- The MAGIC collaboration has completed the
installation of a second telescope, dubbed
MAGIC-II. - MAGIC-II is currently in commissioning phase
until the end of the Summer. - Together, MAGIC-I and MAGIC-II will operate in
stereoscopic mode. - The telescope and its status will be introduced
in the next talk. Here I will focus on main
IFAEs contributions readout and datacenter.
First Light Ceremony, April 2009
24Ultrafast FADCs
- GOAL equip the MAGIC telescopes with ultrafast
(gt2 Gs/s) high-resolution (gt10 bit) digitizers. - Why faster digitizers ?
- -gt Use shower timing information to
discriminate ?-rays from hadrons and muons at low
threshold. - -gt Better discriminate Cherenkov pulses from
night sky noise (and single-phe calibration?) -
25Timing information
gamma candidate
hadrons
muon
Three events of different type showing the energy
(left) and arrival times (right)
It will improve the time resolution to make a
better g/h separation.
26Ultrafast compact FADCs
- How to build them ?
- 1) need a fast and accurate signal sampler -gt
Domino Ring Sampler (DRS) -gt Siena/Pisa - 2) need a fast and efficient event building
and data transmission -gt Pulsar boards S-link
-gt IFAE - -gt Non-trivial instrumental development which has
industrial applications and is under
consideration for future telescopes (CTA)
27Domino Ring Sampler
Domino Ring Sampler developed by Stefan Ritt in
PSI 2-4 GSamples/sec with gt10bit resolution DRS
is integrated into acquisition boards
28MAGIC-II readout general scheme
29Readout Receiver boards
- Interfaces analog optical signals from the
camera with the L1 trigger and the digitization
electronics. - Discriminator programmable threshold, time
delay and width of output pulse. - Online monitoring of individual pixel trigger
rate and temperature. - Generates reference levels for domino
calibration.
30Readout Pulsar boards
- 9U VME board
- Designed for CDF L2 trigger
- 4 mezzanine cards per domino board.
- 2 domino chips per mezzanine
- Each domino (DRS2) chip digitizes 10 channels.
FILAR 4 S-link 360 MB/s
HOLA 160 MB/s
31Readout production
- Initial idea work share on 577 channel
readout - - Siena/Pisa -gt Complete system to upgrade
MAGIC-I - - IFAE -gt Complete system for MAGIC-II
- However, collaboration decided to extend camera
to 1039 pix, so we decided to build only the
readout for MAGIC-II - - Siena/Pisa -gt Mezzanines with DRS chips
- - IFAE -gt Receiver boards with optical
conversion, signal conditioning and
discriminators Data AcQuisition domino
receiver control software cooling racks - Time line to construction
- - Design and prototyping -gt 2006 - 2007
- - Full production and installation -gt 2007 -
2008 - -Commissioning 2008 - 2009.
32Readout production
Crate with optical receiver boards cable
connections to trigger and FADCs (2008)
Receiver Crates (3U Backplane) Inside
water-cooled isolated racks to control temperature
FADC Crates (6U Backplane)
Full readout system in La Palma (2009)
33Readout status
- Currently
- All channels are installed and read out.
- Debugging the readout system.
- Integrating it with the trigger and the other
susbsystems. - Measuring performance of MAGIC-II and MAGIC
system. - First stereo data recorded in the last month.
First events (January 2009)
34MAGIC system datacenter
- With the startup of MAGIC-II, the datacenter will
be officially transfered from Germany to Spain. - During the past 3 years, the groups of IFAE, UCM
and UAB have set up a scaled-down version of new
datacenter at PIC Barcelona. - The datacenter now stores all MAGIC-I data in
tape, makes it available to collaboration and
performs an automatic standard analysis.
35MAGIC system datacenter
- Overall data yield of MAGIC system is 400 Tbyte /
year, i.e. 4 times higher than for MAGIC-I alone. - We want to guarantee for the telescope system
- Data transfer La Palma - PIC.
- Data storages in tape robot.
- Real-time automatic analysis.
- Massive re-analysis on demand.
- Data access to all processing levels.
- Limited resources for private processing of
collaborator members. - Access to the MAGIC data to the whole community
(Virtual Observatory.)
36MAGIC-I CAMERA AND READOUT UPGRADE
37MAGIC-I upgrade
- The collaboration has decided not to build any
more telescopes. - However there are good reasons for one last
upgrade clone the MAGIC-II camera to replace the
old MAGIC-I camera, and accordingly clone the
readout. - UCM and IFAE want to collaborate to build the
clone of the readout.
38Why to upgrade?
- The total area (Field of View) of both cameras
is roughly the same, but the trigger area is 50
larger in MAGIC-II. - For MAGIC-II the whole FOV is homogeneously
equipped with small diameter PMTs, i.e. the
sampling of the showers is significantly better.
MAGIC-II 1039 pixels
MAGIC-I 577 pixels
39Why to upgrade?
- Physical reasons
- Performance in general limited by the worst
telescope the ?-ray acceptance is limited by the
trigger area of MAGIC-I. - This is particularly true for extended and
off-center ?-ray sources.
- Technical reasons
- Camera of M-I is degrading fast in the harsh
mountain conditions and IFAE will have to pay
more and more maintenance. - Current readout of M-I limited to 2 GSps and
impossible to scale up to 1039 pixels.Difficult
to trigger and acquire with both telescopes
because buffer of M-I is too short. - MPI has bought 400 HPDs which can be installed
easily in the new camera (but not in old one)
without interference with M-II.
Off-center ?-ray source
40MAGIC-I readout upgrade
- 2010, 1st semester minor modifications to
design, purchase of electronic components. - 2010, 2nd semester full production.
- 2011, 1st semester full test at UCM.
- 2011, 2nd semester installation commissioning.
- Worth to mention Spanish groups want to build
readout of CTA large telescopes. Domino chip is
one of the two alternatives for readout of CTA.
Expertise in readout production and in this chip
may prove crucial for Spanish groups.
41PROJECT OVERVIEW
42FPA2009-07474 project overview
- Datataking and scientific exploitation.
- Maintenance of IFAE hardware.
- MAGIC-I readout upgrade.
- MAGIC system datacenter.
431. MAGIC III scientific exploitation
- MAGIC-I is right now a highly productive
instrument. - Our investment is now paying off!
- We would like to keep exploiting MAGIC-I, along
with MAGIC-II, in the 3 years of the project.
441. MAGIC III scientific exploitation
- Budget is essentially travel
- Datataking shifts 30 k
- MAGIC meetings 50 k
- Conferences and schools 50 k
- Its an optimal time to make a PhD thesis in
MAGIC - Two PhD grants of FPI program.
452. MAGIC III technical maintenance
- We are responsible for maintenance of all IFAE
hardware in M III camera and readout. - Hardware (5 production cost/yr) 120 k
- Maintenance trips 55 k
- Collaboration common fund 54 k
463. M-I readout upgrade budget
- Readout will again be built along with INFN Pisa.
- We will build an exact clone re-use design and
components. - We know the budget accurately.
- We will build our part with UCM, equal shares
IFAE applies for 303.5 k.
473. MAGIC-I upgrade personnel
- Small design changes, production, test setups,
installation and commissioning at the island all
demand technical manpower - We request a support technician 40 k expense
in the project.
484. MAGIC system datacenter
- Spanish groups (IFAE, UCM, UAB) committed
themselves (via MoU) to contribute the new
datacenter. - However little funding and manpower were
allocated in the last project application.
494. MAGIC system datacenter
- To fully deploy and run the datacenter in the
next years the following resources are necessary - UAB group requests 50 k (see talk in the
afternoon), so IFAE requests the remaining 200
k. - For technical developments and maintenance, we
will need technical manpower - Computing scientist for 3 years 120 k
50Available manpower for FPA2009-07474
Staff researchers engineers (FTE2.5) Postdocs
students (FTE9.5) TOTAL FTE12
51Project total budget
- Personal - Personnel 160 k
- Compl. salariales/salary supplements
0 k - Equipamiento - Equipment 15 k
- Fungible- Consummable 678 k
- Viajes Travel 175 k
- Varios - Others 0 k
-
- TOTAL COSTES DIRECTOS 1027 k
52Outlook
- IFAE has successfully achieved the goal of
participating among the leading institutes in the
construction of MAGIC-I and physics analysis of
the four years of MAGIC-I data. MAGIC-II is now
ready and will start to operate regularly in
stereo mode after the Summer. - IFAE wants to keep its profile as one of the
strongest institutes in extracting physics from
the telescope system. - The MAGIC telescopes will be completed with the
upgrade of the MAGIC-I camera and readout and
will stay as a very powerful and productive
instrument until the advent of CTA in 2015. - IFAE wants to contribute to complete the
instrument by building a clone readout system for
MAGIC-I and deploying the datacenter.
53Backup
54Microquasar Cyg X-1
- Microquasar stellar equivalent of quasar
(active galaxy). - The compact object is a black hole of Mgt13M
- Power released by the (dark) jet is of the same
order or the bolometric x-ray luminosity and 2
orders of magnitude higher than what inferred
from the radio spectrum.
Map in radio extended bow-shock produced by jet
in interaction with interstellar matter.
A natural LHC pumping protons into the galaxy
55Hint of VHE emission
- The source was detected with MAGIC for only one
night in September 2005 (ApJ Lett. 665 (2007)
L51). - First evidence of VHE emision of a stellar mass
black hole. - There is correlation between MAGIC detection and
increase in soft and hard X-rays i.e. when a blob
is ejected in the jet.
56HESS observation of PKS2155 (z 0.116) flare in
2006
- No delay observed ? MQG1 gt 0.52 x 1018 GeV
MQG2 gt 1.4 x 109 GeV
57MAGIC-II sensitivity and angular resolution
58Current FADC vs. DRS
59Manpower request
60Consummables
61The threshold
- We are publishing with a threshold of 90 GeV
- We detect significant signal above 50 GeV
- Understanding our efficiency towards the goal of
50 GeV. A special task force (UHU) has been set
up preliminary physics results at 50 GeV. - Substantial improvement on DM studies and
determination of cosmological constants
62FADC layout
P 1
DRS CRAB
VME
TRIGGER, BUSY
I / O
P 0
DRS CRAB
IPR SCALER
P 2
CTRL
DIGITAL MODULE
DRS CRAB
I / O
P 3
DRS CRAB
S-LINK HOLA
VME Backplane (Configuration)
Pulsar Board (VME 9U) Incl. Digital modules
Auxiliary Board
63MAGIC-II readout production