1. CANGAROO-III and beyond
Masaki Mori*
for the CANGAROO team
*ICRR, The University of Tokyo
Pre-ICRC workshop: New Generation Cherenkov Imaging Telescopes
Aug 1-2, 2005, Mumbai, India

2. “CANGAROO” =
Collaboration of Australia and Nippon for a GAmma Ray Observatory in the Outback
Woomera, South Australia

3. CANGAROO team University of Adelaide Australian National University
Ibaraki University
Ibaraki Prefectural University
Konan University
Kyoto University
STE Lab, Nagoya University
National Astronomical Observatory of Japan
Kitasato University
Shinshu University
Institute of Space and Astronautical Science
Tokai University
ICRR, University of Tokyo
Yamagata University
Yamanashi Gakuin University
15 universities from Austraila and Japan, about 40 people

4. Brief history of CANGAROO
1987: SN1987A
1990: 3.8m telescope
1990: ICRR-Adelaide Physics agreement
1992: Start obs. of 3.8m tel.
1994: PSR
1998: SNR1006
1999: 7m telescope
2000: Upgrade to 10m
2001: U.Tokyo-U.Adelaide agreement
2002: Second and third 10m tel.
2004: Four telescope system

5. (BIcentennial Gamma RAy Telescope)
Why Woomera?
NZ: too wet, not many clear nights
Woomera:
Former rocket range and prohibited area…infra-structure and support
Adelaide group was operating BIGRAT
ELDO rocket Launch site in ’60s
BIGRAT
(BIcentennial Gamma RAy Telescope)

6. CANGAROO-II telescope
Upgraded in 2000 from 7m telescope completed in 1999
114 x 80cm CFRP mirror segments in parabola (first plastic-base mirror in the world!)
Focal length 8m
Alt-azimuth mount
552ch imaging camera
Charge and timing electronics
(March 2000)
Tanimori et al., ICRC 1999

7. CFRP mirror & tuning system
80cm, 5.5kg
Kawachi et al., Astropart.Phys. 14, 261 (2001)

8. CANGAROO-II camera 3 FOV R4124UV (Hamamatsu) 0.115 pixel Lightguide
16PMTs/module

9. CANGAROO-II Electronics

10. CANGAROO-II & -III

11. Woomera: 2004 March
The observatory is near Woomera, at the southern edge of Australian desert. Now we completed four ten meter telescope as planned.
T T T3 T1

12. Basic specifications of telescopes
Location:
3106’S, 13647’E
160m a.s.l.
Telescope:
114 80cm FRP mirrors
(57m2, Al surface)
8m focal length
Alt-azimuth mount
Camera:
T1: 552ch (2.7 FOV)
T2,T3,T4: 427ch (4 FOV)
Electronics:
TDC+ADC T2
The telescope reflector consists of cm diameter small spherical mirrors made of fiber-reinforced plastic with aluminium surface.
The first telescope has a 2.7degree field-of-view camera, and the rest three have 4 degree cameras.
Mori et al., Snowbird WS (1999)

13. GFRP mirrors and tuning system
Tuning using star images via a CCD camera
2.0
Before tuning
After tuning
Ohishi et al., ICRC 2003

14. Spot size T4 Y (vertical) X (horizontal) 0.7
Point Spread Function (FWHM)
T1: 0.20
T2: 0.21
T3: 0.14
T4: 0.16
Y (vertical)
(measured at construction time)
This is an image of a star focused by our reflector at the focal plane. The full-width-at-half-maximum size of the point spread function is about 0.2 degree. This value is not so good but still less than the size of one pixel of our camera.
Image of a star on camera observed by a CCD camera
X (horizontal)

15. CANGAROO-III camera R3479 (Hamamatsu) Lightguide (T1/T234) T1 T2,T3,T4
FOV 3 4
Num.of pixel
552
427
Weight
~110kg
Size of PMT ½” ¾”
Pixel arrangement
square
hexagonal
HV polarity
negative
positive
HV supply unit
1ch/16 PMTs
1ch/1 PMT

16. PMT gain uniformity and linearity
Kabuki et al., Nucl. Instr. Meth. A500, (2003)

17. Lightguide design Winstone cone cross section
Efficiency vs. incident angle
Kajino et al., ICRC2001

18. High voltage control & monitor

19. Camera calibration Blue LED flasher at the reflector center
Blue LED flasher in the camera box
Patterned screen
Yamaoka et al., ICRC2003

20. CANGAROO-III Electronics (1)
Kubo et al., ICRC 2003

21. CANGAROO-III Electronics (2)
Discriminator and summing module (DSM)
Trigger logic

22. CANGAROO-III Electronics (3)
Single photoelectron spectrum measured with DSM and ADC
ADC linearity

23. Telescope control Telescope control unit Driving control PC
GPS
Position data (every 100ms)
Position command (alt-azimuth)
RS-232C
Driving control PC
Remote command/position data/NTP
Hayashi et al., ICRC2003
Local area network

24. Star position error observed by a CCD camera
Star tracking
Star position error observed by a CCD camera T3
PMT size
RMS deviation
0.013 degree
CCD Y-axis (degree)
Here are some data showing the performance of our telescopes. This plot shows errors of star positions observed by a CCD camera with a camera lens set at the center of the reflector. It means our tracking accuracy is less than 1 arc miniute and is well within the size of one pixel of our camera.
Hear Kiuchi’s talk!
CCD X-axis (degree)

25. Construction of CANGAROO-III
: Observation start
: Expansion to 10m
: Observation
: Tuning T1 T2 T3 T4
2000
1999
2001
2002
2003
2004 3 11 12 7 1 6
4-telescope stereo
Here is our history of construction of telescopes. The first one is operating in 10m mode for 5 years. The four-telescope array started observation in March last year.

26. Sample of 4-fold stereo events
Here are some samples of 4-fold stereo events. Color scale shows ADC counts which is proportional to number of electrons detected in camera pixels. For the first half year our telescopes were triggered independently and coincident events are searched for using the GPS timing information and common event numbers. Then we introduced a global trigger system which I will mention later.
Data: 2004 March

27. Global trigger system d Telescopes Telescopes Coincidence
Before: “software trigger”
Each telescopes triggered independently
Now: “hardware stereo”
Requires at least 2 telescopes
If no coincidence  Reset
Dead time 1/100
100m
t=d/c < 500ns
variable
650ns
Opt.fiber
150m
Telescopes
Telescopes
Opt.fiber
Turnaround
~2.5s
Trigger
Wait time
~5s
Trigger
Event number
Coincidence

28. Effect of global triggers
without global trigger
muon
with global trigger
hadron
with global trigger
without global trigger
Length/size
Muon events are removed!

29. Beyond CANGAROO-III In the near future In the long range
Improvement of old T1 and others
In the long range
No unified plan yet…
Started brainstorming, technical and physical considerations…

30. Large reflector/high altitude
Where should we go?
Wide FOV camera
Lower Energy
Wider coverage
Large reflector/high altitude
Higher Energy
Large effective area
Higher sensitivity

31. A case study: array of telescopes
How to achieve large effective area in modest cost?
Large span array with wide cameras?
SPAN
Yoshikoshi et al. Paleiseau WS (2005)

32. Lateral distribution of light
Tail is extended beyond 150m!

33. Array span vs. effective area
6 FOV camera
Gamma-ray energy: 100 GeV, 1 TeV, 10 TeV

34. Summary
CANGAROO-III is a system of 10m imaging Cherenkov telescope build by Japanese-Australian collaboration.
We have been carrying out 4-telescope stereo observations of sub-TeV gamma-rays since 2004 March. Now we have incorporated a global trigger system to reduce muons.
We are studying the next-generation telescopes. One option could be a large-span array of telescopes to increase the effective area.
In summary, we are carrying out 4-telescope stereo observations since 2004 March.
Stereo characteristics are under investigation and analysis is underway.
Recent results with the first 10m telescope were given for the Galactic center and a pulsar binary PSR B /SS2883.

35. End

36. Stereo observation Target 2 distribution Intersection point (qx, qy)
Angular resolution
0.25deg  0.1 deg
Energy resolution
30%  15%
Better S/N (no local muons)
2 distribution
Intersection
point
(Simulation)
Target
Entries/bin
Now I turn to our stereo analysis. The procedure is basically the same as that developed by the HEGRA team and I do not go into detail. Square of the angular distance of the intersection point of image axes from the target point, or theta squared, has a peak toward zero if there is a gamma-ray signal.
(qx, qy)
2 = x2+ y2
2 [deg2]

37. Unfortunate situation for the Crab
Showers from the Crab
The oldest T1 has higher energy threshold and bad efficiency for stereo observation
Only T2/T3/T4 are used for stereo analysis
Stereo baseline becomes short for the Crab observation at large zenith angles
Now I have to explain our unfortunate situation for observations of the Crab in CANGAROO-III. The 5-year-old T1 has a higher energy threshold and stereo efficiency is bad, so currently we do not use its data for stereo analysis. Then, the stereo baseline to observe the Crab at large zenith angles, around 55 degree, become very short.

38. Large zenith angle observation of the Crab
Higher energy threshold ~1TeV
Bad intersection accuracy
Narrower
Far coresmall anglebad accuracy
h30
That makes our accuracy to determine intersection points of image axes very bad. First we restricted events whose axes opening angle between 15 degree and 165 degree.
Entries/bin
h60
Accept 15<<165 only  

39. Crab signal (1) Preliminary Preliminary (simple square cuts)
Team “A”
(simple square cuts)
Nov 2003 On
Preliminary
Preliminary
Off
Entries/bin
(On-Off)/bin
This is the theta squared plot for on- and off-source. The excess of events around zero shows a gamma-ray signal. From its shape we can estimate our angular resolution.
2 [deg2]
2 [deg2]
Sigma : 6.19
Excess : 25842 event
Angular Resolution : 0.16 (HWHM)
T2 & T3
ON 7.5hr
OFF 7.0hr

40. Crab signal (2) Significance map Differential flux Preliminary
Team “A”
Significance map
Differential flux
Preliminary
Preliminary
This is the significance map of data around the Crab nebula showing the consistency of its position with the Crab nebula. The right panel is the differential energy spectrum which is consistent with that given by the HEGRA group.
Angular resolution for the Crab (h~35)
~0.17 (RA) / 0.14  (Decl)