Development of cold-slumping glass mirrors for imaging Cherenkov telescopes

Development of cold-slumping glass mirrors for imaging Cherenkov telescopes D. Vernani1, R. Banham1, O. Citterio1, F. Sanvito1, G. Valsecchi1, G. Pareschi2, M. Ghigo2, E. Giro3 M. Doro4, M. Mariotti4 1 2 3 Media Lario Technologies, Località Pascolo – 23842 Bosisio Parini (LC), Italy. INAF / Brera Astronomical Observatory, via E. Bianchi 46 – 23807 Merate (LC), Italy INAF / Osservatorio Astronomico di Padova, Vicolo Dell’Osservatorio 5, 35122, Padova 4 Università and INFN Padova, Via Marzolo 8, 35131 Padova (PD), Italy ABSTRACT The development of lightweight glass mirrors manufactured via cold-slumping technique for Imaging Atmospheric Cherenkov Telescope is presented. The mirror elements have a sandwich-like structure where the reflecting and backing facets are composed by glass sheets with an interposed honeycomb aluminum core. The reflecting coating is deposited in high vacuum by means of physical vapor deposition and consists of aluminum with an additional protective layer of SiO2. The mirror fabrication and environmental qualification by accelerated ageing, thermal cycling and coating adhesion are presented together with the optical performances measured as angular resolution and reflectivity obtained on spherical, 1 squared meter mirror prototypes. Keywords: Advance Optics, Segmented Mirror, Sandwich Mirror, Cherenkov Telescope 1. INTRODUCTION The emission of the gamma-ray by cosmic sources was predicted by scientists long before the real experimental detection. Theoretical works by E. Feenberg and H. Primakoff in 1948 [1], S. Hayakawa in 1952 [2], and P. Morrison in 1958 [3] had led scientists to believe that a number of different processes, which were occurring in the universe, would result in gamma-ray emission. The first true astrophysical gamma-ray sources were solar flares, which revealed the strong 2.223 MeV line predicted by P. Morrison. Significant gamma-ray emission from our galaxy was first detected in 1967 by the gamma-ray detector aboard the OSO-3 satellite [4]. Perhaps the most spectacular discovery in gamma-ray astronomy came in the late 1960s and early 1970s from a constellation of defense satellites which were put into orbit for a completely different reason. Detectors on board the Vela satellite series, designed to detect flashes of gamma-rays from nuclear bomb blasts, began to record bursts of gamma-rays not from the vicinity of the Earth, but from deep space [5]. Today, these Gamma Ray Bursts (GRB) are seen to last for fractions of a second to minutes and then fading after briefly, constituting one of the prevalent topics of modern astrophysics. Very High Energy (VHE) gamma rays, with photon energies over ~30 GeV, can also be detected by ground based experiments. In fact, such high energy photons interact high up in the atmosphere and generate an air shower of secondary particles. These particles emit the so-called Cherenkov light, a faint blue light. The Cherenkov light illuminates an area of about 250 m diameter on the ground and a telescope located somewhere within the light will detect the air shower, provided that its mirror area is large enough to collect enough photons. The image obtained with the telescope shows the track of the air shower, which points back to the celestial object where the incident gamma-ray originated. The Crab Nebula, a steady source of so called TeV gamma-rays, was first detected in 1989 by the Whipple Observatory [6]. The Imaging Atmospheric Cherenkov Telescope (IACT) is the technique that currently achieves the highest sensitivity in the VHE gamma-ray observations. Gamma-ray astronomy has experienced a major breakthrough with the impressive astrophysical results obtained mainly by the current generation of Cherenkov experiments like HESS [7]-[8], VERITAS [9], MAGIC [10], and CANGAROO [11]. IACT gamma-ray astronomy observations are, for the moment, limited by non-gamma ray backgrounds at lower energies, and, at higher energy, by the number of photons that can be detected. Larger area detectors and better background suppression are essential for progress in the field. Some new experiments, like HESS II [12] and MAGIC II [13], have been recently undertaken aiming at improving the Advanced Optical and Mechanical Technologies in Telescopes and Instrumentation, edited by Eli Atad-Ettedgui, Dietrich Lemke, Proc. of SPIE Vol. 7018, 70180V, (2008) · 0277-786X/08/$18 · doi: 10.1117/12.790631 Proc. of SPIE Vol. 7018 70180V-1 2008 SPIE Digital Library -- Subscriber Archive Copy capability of the existing IACT observatories. The impressive physics achievements obtained with the present generation instruments has triggered the initiative of astrophysicists to consider future ground-based gamma-ray observatories, like CTA [14] and AGIS [15], consisting in large array of telescopes. All of these Cherenkov telescopes use optical systems consisting of large segmented mirrors, focusing the Cherenkov light onto photon detectors made of photomultiplier to resolve the image of the air shower. Task of the optical system of IACTs is to collect Cherenkov light and to focus it onto the detector. The point spread function should ideally be smaller than the pixel size over the entire field of view. Whereas the imaging performance of IACTs is modest compared to normal astronomical telescopes, the cost-effective design and the lightweight of the mirrors of large Cherenkov telescopes is non trivial. Different technologies have been adopted so far for the production of the Cherenkov segmented mirrors. The different technologies can be divided into three main groups: ƒ ƒ ƒ aluminized ground-glass mirrors manufactured with standard technique starting from raw blanks; composite sandwich structure mirrors manufactured via direct machining of each individual piece; composite sandwich structure mirrors manufactured via replication process from a mould. Ground-glass mirror solution has been often preferred (e.g. HEGRA [16], CAT [17], HESS, VERITAS) primarily because of its technical maturity, but at cost of a quite long time of production. Moreover, the ground-glass mirrors are quite heavy translating into increasing cost and complexity for the telescope mechanical structure. Thanks to the limits of the ground-glass technology, the idea to make use of lightweight mirror consisting of composite sandwich structure came to pass since the very beginning of IACT astronomy. In sandwich construction, membranes (such as sheet steel, aluminum, glass, or plastic) are bonded to both side of a core material. This type of construction is widely utilized in products ranging from doors and tables to aircrafts, boats and satellites and is characterized by high strength-to-weight. The first Cherenkov Telescope employing composite sandwich mirrors had been the MARK 3 experiment, for which a replication process had been developed by making use of aluminum honeycomb core and Alanod® face sheets [18]. The CANGAROO III telescope, instead, adopts composite sandwich mirrors consisting of rigid foamed core pinched by Fiber Reinforced Plastic sheets; in this case the structure is placed on a mould and curved in an auto-clave for shape replication [19]. Also the MAGIC mirrors are composite structure, in this case consisting of an aluminum face sheet premachined to spherical shape and glued to an aluminum honeycomb inside a thin aluminum box making up the raw blank; each individual raw blank is subsequently polished by diamond milling [20]. For the MAGIC II telescope 236 one squared meter mirrors have been manufactured, the first 136 mirrors have been realized with an improvement of the technique already adopted for the MAGIC telescope [21], whereas the remaining 100 mirrors have been manufactured with the cold-slumping technique here described (the status of the 100 mirrors for the MAGIC II telescope is presented in a different paper of this conference [22]). The main goal of this paper is to present the optical performance and the qualification test performed on prototype coldslumped glass mirrors. The following chapters describe the general design consideration, the mirror fabrication, the numerical finite element analysis, and the environmental qualification together with the angular resolution and the reflectivity performances. 2. SEGMENTED MIRROR FABRICATION VIA COLD SLUMPING The cold slumping technique makes use of the vacuum suction to mechanically bend the reflecting membrane of a composite mirror. The possibility to use glass sheets in sandwich structures was first investigated for Solar concentrators application [23]. If the radius of curvature of the optics is high and the thickness of the glass sheet is sufficiently low, the sheet can be conformed to the shape of the master by means of vacuum action. The mirror elements here proposed for Cherenkov Telescope application have a sandwich like structure where the reflecting and backing glass sheets are bonded to both sides of an aluminum honeycomb core. A schematic illustrating the construction of the sandwich structural mirror is illustrated in Fig. 1. The fabrication process is composed by the following steps: 1. Spherical curvature of the surface of an aluminum mould is obtained by diamond milling. Diamond milling is used in order to achieve the best shape accuracy possible without the need for further polishing (see also Fig. 2 on the left); 2. the shape of the mould is replicated by the reflecting glass sheet via the cold slumping process. A backing glass sheet is assembled with an interposed aluminum honeycomb core element giving the proper rigidity; Proc. of SPIE Vol. 7018 70180V-2 Fig. 1. Sketch of the cold-slumping mirror manufacturing process 3. the connection of the parts is achieved through epoxy resin structural adhesive bonding with curing under elevated temperature while maintaining the vacuum suction; 4. the glass sheets adopted are floating glass available on the market with a very good roughness and do not require any polishing step. The process ensures on the reflecting glass sheet the required shape accuracy after separation from the master and t3he preservation of the starting surface roughness of the glass sheets; 5. the reflecting coating is deposited after the manufacturing of the sandwich structure by means of physical vapor deposition in a dedicated high vacuum chamber. The reflecting glass sheet is coated in order to provide a high reflectivity at wavelengths in the range from 300 to 600 nm. Aluminum coating provides the best reflectivity at these wavelengths, especially in the range of short wavelengths (300 to 450 nm) that contains most intensity of the Cherenkov light. To avoid oxidation of the aluminum layer, a protective coating of quartz is also applied. 6. Sealing of the sandwich structure borders is assured by a silicon based sealant. The edges of the sandwich have an external plastic PVC rim. This solution assures higher rigidity and mechanical protection of the mirror corners. All the materials within process are off-the-shelf and the higher cost is one time expenditure relevant to the manufacturing of the mould. Moreover the mould is not subjected to significant degradation during the process. In the following Fig. 2 the pictures of the master and of one prototype mirror are shown. Fig. 2. Left: mould shaped by means of diamond milling; right: prototype mirror replicated via cold slumping. Proc. of SPIE Vol. 7018 70180V-3 3. SEGMENTED MIRROR FINITE ELEMENT ANALYSIS The shape accuracy and the strength analysis of the glass sandwich structure have been performed under different load cases with the following boundary condition: ƒ ƒ ƒ Sandiwch structure dimension 985 mm x 985 mm; Radius of curvature of the spherical shape 34 m; Sandwich structure is composed of the following layers: − − − − ƒ − reflecting sheet 1.7 mm thick (floating glass) bonding 0.2 mm thick (epoxy resin) aluminum honeycomb 20 mm thick bonding 0.2 mm thick (epoxy resin) backing sheet 1.7 mm thick (floating glass) 3 rigid supports with translations locked and rotations free. The three supports are assumed to be at a radius equal to 2/3 of the mirror structure side. Rigid cross beams are implemented in correspondence of the supports in order to reproduce stiffness of inserts (diameter 80 mm). The circular shape is recommended for the inserts in order to limit concentrated stress loads in facings. Finite Element Model sandwich structure mass is 9.4 kg. Load cases considered in the finite element analysis are defined in the following Table 1. The operational temperature range is between 0°C and 30°C, whereas the survival condition is between -20°C and 60°C. The reference temperature is 20°C. Table 1. Load cases for the Finite Element Analysis Load case Description Notes LC1a Normal gravity 1g along Z LC1b Lateral gravity 1g along X LC2a Operational hot case (uniform temp. 30°C) Uniform temp. change +10°C LC2b Operational cold case (uniform temp. 0°C) Uniform temp. change -20°C LC2c Survival hot case (uniform temp. 60°C) Uniform temp. change +40°C LC3a Temperature gradient 1°C Reflecting skin 20.5°C Backing skin 19.5°C The results of the structural analysis in terms of shape accuracy under the different load cases are summarized in the following Table 2. Note that PTV and RMS errors are calculated w.r.t. unreformed shape and considering vertical displacements. Best fit is performed only w.r.t. piston effect. The stresses have been calculated as equivalent von Mises stresses. Adhesive shear stress is calculated from shear stress in core, by scaling it according to reduced section area. The results of the stress analysis are summarized in the following Table 3. Proc. of SPIE Vol. 7018 70180V-4 Table 2. Shape accuracy under the different load cases Vertical displacement Load case PTV error RMS error Piston RMS error w/o [µm] [µm] [µm] Piston [µm] Slope error PTV [mrad] LC1a Normal gravity 15.9 5.0 -3.9 3.1 0.033 LC1b Lateral gravity 0.3 0.1 0.0 0.1 0.001 LC2a Hot case (30°C) 2.4 0.9 -0.6 0.6 0.005 LC2b Cold case (0°C) 4.8 1.8 -1.2 1.2 0.011 LC3a Gradient 1°C 8.5 2.0 -0.6 1.9 0.024 Table 3. Stress analysis under the different load cases Margin of safety (2) Stress [MPa] Load case Skin Eq. Von Mises (1) Adhesive shear LC1a 0.59 (0.58) LC1b Core shear Skin Eq. Von Mises Adhesive shear W (XZ) L (YZ) 0.20 0.013 0.015 21.6 0.13 (0.13) 0.00 0.000 0.000 LC2b 6.2 (5.6) 0.07 0.006 LC2c 12.3 (11.2) 0.13 LC3a 0.08 (0.08) Max allowable 20 Core shear W (XZ) L (YZ) 96.4 71.3 106 102 8645 4699 16066 0.003 1.16 287 156 535 0.012 0.006 0.08 73.5 77.3 266.8 0.00 0.000 0.000 166 6834 4699 8032 29 1.41 2.41 - - - - Notes: 1) First principal stress is also shown (in brackets) 2) Safety factor of 1.5 included in margins of safety Proc. of SPIE Vol. 7018 70180V-5 4. QUALIFICATION TEST RESULTS In this session the activities performed for the environmental qualification of the cold-slumped mirrors for Cherenkov Telescopes are presented together with their performance in terms of angular resolution and reflectivity. Some qualification tests have been performed at the level of samples (smaller with respect to the final dimension but fully representative of the manufacturing process) and some other test have been performed on 1 squared meter prototype. In some cases the tests have been performed both on the samples and on the prototype mirrors. In the Table 4 the list of the qualification test is reported with the indication whether the test has been performed on small scale samples or on real final dimension prototypes. Table 4. Summary of the qualification tests with the indication (T) whether the test has been performed on small samples or on real dimension prototypes Qualification Test (Type-Method) Test on Samples Test on Prototypes 1 Angular Resolution T T 3 Reflectivity T T 2 Angular Resolution after Thermal Cycling (-20°C / +60°C) 4 Reflectivity (before and after Weathering Test) T 5 Reflectivity (before and after Salt Fog Test ) T 6 Coating Adhesion test T 7 Sealing Test T T T Before starting the qualification tests, the angular resolution and the reflectivity of the prototypes mirror have been characterized with two different experimental set-ups. For the Point Spread Function (PSF) measurement, the mirror was placed at a distance of ~34 m (the assumed radius of curvature of the mirror) in respect to a monochromatic red diode. The diode is simulating a point-like light source. A white sheet of paper was placed into a paper box and located close to the diode, at the same distance to the mirror. The distance from the mirror to the diode and reflecting plane was adjusted until the reflected spot size had its minimum. The reflected spot was imaged with a 16bit CCD camera. '-.- 20 —. 2000 —- 1500 1000 -20 -20 -15 -10 -5 0 5 10 15 20 mm Fig. 3. Point Spread Function (PSF) of a cold-slumped glass mirror at distance of ~34m. The circle containing 80% of the light has a radius of ~7.4 mm that corresponds to angular radius of ~0.22 mrad (MPI Munich credits). Proc. of SPIE Vol. 7018 70180V-6 The spot diagram of a representative cold-slumped mirror is reported in Fig. 3, where the circle containing 80% of the light has a radius of ~7.4 mm that corresponds to an angular radius of ~0.22 mrad. The focal length of the cold-slumped mirrors is normally slightly shorter than the focal length of the mould from which they are replicated. This aspect of the process, although not expected, is very repeatable and hence the focal length of the mould can be designed in order to take it into account. An example of the spread in focal length of the cold-slumped mirrors can be found in a separate paper talking about the production of the 100 mirrors for MAGIC II telescope [22]. For the measurement of the mirror reflectivity the portable IRIS 908RS2 instrument was used. This instrument allows to measure two components of the reflectivity, the direct and the diffuse reflected component, on several position on the mirror reflecting surface. An image of the portable device for the reflectivity measurement is reported in Fig. 4 (on the right). The device is operating at four different wavelengths. The four laser diodes have the following characteristics: − B: λ = 470 nm (FWHM = 30 nm) − G: λ = 530 nm (FWHM = 40 nm) − R: λ = 650 nm (FWHM = 30 nm) − IR: λ = 880 nm (FWHM = 80 nm) The laser diodes focus their beam of light at an inclination angle of 45° in respect to the sample surface. Four detectors are mounted at 0°, +2°, -15° and -45°. The later three are used to determine the scattering component of the reflected light. The maximal relative error of the instrument and therefore the measurement accuracy is 0.5%. The repetitiveness of the reflectivity measurement is 0.1% for 20 measurements. The measured area is 10mm in diameter. Fig. 4 (on the left side) shows the measurement of the mirror reflectivity performed on a representative 1 squared meter prototype glass mirror. The reflectivity is well above 80% and peaks in the blue region (where the Cherenkov light emission is higher). The scattering amount of the cold-slumped mirror is typically lesser than 0.1%. 94 92 90 Reflectivity (%) 88 86 84 82 80 78 76 460 480 500 520 540 560 580 Wavelength (nm) 600 620 640 660 Fig. 4. Reflectivity measurement at different wavelength of a typical prototype cold-slumped mirror Angular Resolution After Thermal Cycling The measurements of the PSF of the prototype mirrors and the radius of curvature measurements have been performed before and after 5 thermal cycles. These tests aimed at the verification of the maintenance of the optical performance of the mirror in the survival condition. Each thermal cycle has been performed passing quickly from 20°C to -20°C (plateau of 12 hour) and then from -20°C to 60°C (plateau of 4 hours). The temperature profile is reported below in Fig. 5. After each cycle the radius of curvature and the angular resolution (PSF) have been measured showing that no changes occurred with respect to the starting parameter before thermal cycling. Proc. of SPIE Vol. 7018 70180V-7 50 40 Temperature °C 30 20 10 0 -10 -20 0 2 4 6 8 10 12 Hour 14 16 18 20 22 Fig. 5. Temperature curve of the thermal cycles performed on the prototype mirrors Reflectivity Before and After Weathering Test The reflectivity of 3 samples with dimension 200 mm x 200 mm have been measured before and after accelerated aging test by means of IRIS 908RS2 portable instrument. The accelerated aging test (or weathering test) lasted 42 days with continuous variation of temperature profile, relative humidity profile and ultraviolet radiation power (UVA). The temperature ad humidity profiles are plotted in Fig. 6. The UVA power irradiating the samples had been switched during the test between the following values: − − 2.2 mW/cm2 with the lightening of 5 lamps at distance of 65 cm; 4.4 mW/cm2 with the lightening of 10 lamps at distance of 65 cm. Wssthsrlng Tsut . Tsn,psrsturs Protlis Weathering Test - Relative Humidity Profile 80 00 40 a 7IIIItttIII±IrII 70 --- -I : 46 20 46 I III III -20 46 I! I I LL I II . I. - -—- -I — o Ii 12 13 14 5 16 1 I 19 20 I 22 22 24 Time tht -40 rime oni Fig. 6. Temperature profile (left) and relative humidity profile (right) during weathering test on samples The reflectivity measurements are reported in the following Table 5. As it can be seen from the table the reflectivity of the mirrors is not significantly affect by the weathering conditions. Table 5. The reflectivity of 3 samples measured before and after accelerated aging test Wavelength Sample 1 Before Test Sample 1 After Test Sample 2 Before Test Sample 2 After Test Sample 3 Before Test Sample 3 After Test 470 nm 88.73 % 88.09 % 88.45 % 87.97 % 88.63 % 87.86 % 530 nm 84.42 % 83.66 % 84.07 % 83.56 % 84.25 % 83.84 % 650 nm 80.50 % 80.27 % 80.49 % 80.18 % 80.67 % 80.3 % Proc. of SPIE Vol. 7018 70180V-8 Reflectivity Before and After Salt Fog Test In order to assess the corrosion resistance of the cold-slumped mirror coating, a Salt fog test have been performed on some representative samples. Salt fog test produces an accelerated corrosive attack in order to predict the coating suitability in use as a protective finish on corrosive environments. The Salt fog test has been performed with an atomized fog of water having a high salt content (higher than 5%) for the duration of 24 hours. Even if there is no clear correlation between 24 hours of exposure with a 5% solution to a long period of exposure in an actual corrosive atmosphere, the test has been performed without showing signs of corrosion on the samples. Also the reflectivity of the coating has not been affected by the Salt fog test as measured with the IRIS 908RS2 portable instrument. In Fig. 7 the experimental set-up is depicted showing the samples installed in the chamber (on the left) also during the execution of the test (on the right). Fig. 7. Samples installed in the chamber for Salt Fog Test (left) and during the execution of the Salt Fog Test (right) Coating Adhesion Test The adhesion of the coating has been checked by a peel-off test performed both on samples and on prototype mirror. The processes of aluminization and protective quartz layer deposition are performed under high vacuum (~10-6 mbar) in a dedicated chamber via physical vapor deposition; the reflecting surface is cleaned through Ion etching before the deposition. Pressure sensitive tape has been applied and removed over different areas in the coating. The areas have been subsequently inspected for removal of coating from the substrate giving no indication of detachments. Coating adhesion tests have been performed also after Weathering and Salt Fog Test performed on representative samples without giving indication of detachment of the coating Sealing Tightness Test For pressure exchange between sandwich mirror interior and external environment, required by the sealing tightness test, a venting hole has been provided on one side of some prototype mirrors. Sealing tightness of prototype mirror has been tested by means of a vacuum pump. The internal pressure has been reduced from ambient to 0.2 bar and the vacuum circuit has been therefore closed. No leakage has been detected within 15 minutes. As additional test, some mirror prototypes have been dipped completely into water for the duration of 24 hours. The weight of the mirrors before and after the test was exactly the same giving indication that no penetration of water inside the sandwich structure occurred. Proc. of SPIE Vol. 7018 70180V-9 5. CONCLUSION Cost-effective design of the IACT segmented mirrors and its production rate represent key ingredient for the success of the next generation instrument With cold slumping technique it is possible to manufacture lightweight (9.5 kg/m2) mirrors in quite short time (e.g. 5 mirrors per day if 5 masters are available), with optical quality within the requirements of Cherenkov Telescopes and by using materials available off-the-shelf. Although the development study has been performed on spherical shape mirrors, also parabolic profiles are possible once a master with the right shape is available. A number of environmental test have been successfully performed and the finite element analysis does not give indication of criticalities within Cherenkov Telescope boundary conditions. More than 100 mirrors have been produced with the cold slumping technique for MAGIC II telescope with commissioning foreseen for the next 19th September. ACKNOWLEDGMENTS The authors thank HESS and MAGIC collaborations for supporting the qualification activities. MPI Munich is acknowledged for the optical measurements performed on some prototype mirrors. Special thanks go to Riccardo Ghislanzoni (MLT) for the 3D model and to Josef Eder for the Finite Element Analysis. The Diamond Milling of the aluminum mould has been performed by LT Ultra Company (Aftholderberg, Germany). The Weathering tests have been performed by TUV Italy. The aluminum and quartz coatings have been performed by ZAOT Company (Vittuone, Italy). REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] E. Feenberg and H. Primakoff, “Interaction of cosmic ray primaries with sunlight and starlight”, Phys. Rev. 73, 449 (1948). S. Hayakawa, “Propagation of the Cosmic Radiation through Intersteller Space”, Progress of Theoretical Physics Vol. 8 No. 5, 571-572 (1952). P. Morrison, “Solar Origin of Cosmic Ray Time Variations,” Proceedings of the Fifth International Congress on Cosmic Radiation, 305 (1958). W.L. Kraushaar, G.W. Clark, G. Garmire, “Preliminary results of gamma-ray observations from OSO-3,” Can. J. Phys. Vol. 46, S414 - S418 (1968). I.B. Strong, R.W. Klebesadel, R.A. Olson, “Preliminary Catalog of Transient Cosmic Gamma-Ray Sources Observed by the VELA Satellites,” Astrophysical Journal Vol. 188, L1 (1974). T.C. Weekes, M.F. Cawley, D.J. Fegan, K.G. Gibbs, A.M. Hillas, P.W. Kowk, R.C. Lamb, D.A. Lewis, D. Macomb, N.A. Porter, P.T. Reynolds, G. Vacanti, “Observation of TeV gamma rays from the Crab nebula using the atmospheric Cerenkov imaging technique,” Astrophysical Journal Vol. 342, 379-395 (1989). W. Hofmann and the HESS Collaboration, “The High Energy Stereoscopic System (HESS) Project,” AIP Conference Proceedings Vol. 515, 500 (2000). K. Bernlöhr and the HESS Collaboration, “Optical system of the H.E.S.S. imaging atmospheric Cherenkov telescopes. Part I: layout and components of the system,” Astroparticle Physics Vol. 20 Issue 2, 111-128 (2003) H. Krawczynski and the VERITAS collaboration, “The VERITAS Gamma-Ray Observatory - Status and Recent Results,” American Astronomical Society AAS Meeting 212, (2008). E. Lorenz and the MAGIC Collaboration, “The MAGIC Telescope Project,” AIP Conference Proceedings Vol. 515, 510 (2000). M. Mori et al., “The CANGAROO-III Project,” AIP Conference Proceedings Vol. 515, 485 (2000). D. Horns and the H.E.S.S. Collaboration, “H.E.S.S.: Status and future plan,” Journal of Physics: Conference Series, Volume 60, Issue 1, pp. 119-122 (2007). F. Goebel and the MAGIC Collaboration, “Status of the second phase of the MAGIC telescope,” Contribution to the 30th ICRC, (2007). G. Hermann, W. Hofmann, T. Schweizer, M. Teshima and the CTA consortium, “Cherenkov Telescope Array: The next-generation ground-based gamma-ray observatory,” Contribution to the 30th ICRC, (2007). J. Buckley, “The Advanced Gamma-ray Imaging System (AGIS),” American Physical Society APS Meeting and HEDP/HEDLA Meeting, (2008). A. Daum et al., “First results on the performance of the HEGRA IACT array,” ASTROPARTICLE PHYSICS Volume 8, 1-11 (1997). Proc. of SPIE Vol. 7018 70180V-10 [17] [18] [19] [20] [21] [22] [23] A. Barrau et al., “The CAT imaging telescope for very-high-energy gamma-ray astronomy,” Nucl. Instrum. Methods Phys. Res. Vol. 41, 278 – 292, (1998). I. Carstairs, P.M. Chadwick, N.A. Dipper, E.W. Lincoln, T.J.L. McComb, “The University of Durham new VHE gamma ray telescopes,” Proceedings of the NATO Advanced Research Workshop Durham, England, Aug. 11-15, (1986). M. Ohishi et al., “Status of CANGAROO-III,” The Universe Viewed in Gamma-rays - University of Tokyo Workshop, (2002). D. Bastieri for the MAGIC Collaboration, “The Mirrors for the MAGIC Telescopes,” 29th International Cosmic Ray Conference Pune 00, 101–106, (2005). D. Bastieri on behalf of the MAGIC Collaboration, “The reflecting surface of the MAGIC-II Telescope,” ARXIV Contribution to the 30th ICRC, (2007). G. Pareschi, R. Canestrari, O. Citterio, M. Ghigo, R. Banham, G. Valsecchi, D. Vernani, E. Giro, M. Doro, M. Mariotti, “Glass panels by cold slumping to cover 100 m2 of the MAGIC II Cherenkov telescope reflecting surface,” SPIE Proceedings 7018, (2008). R.B. Diver, J.W. Grossman, “Sandwich Construction Solar Structural Facets,” ASME International Solar Energy Conference, Maui, HI, Apr. 11-14, (1999). Proc. of SPIE Vol. 7018 70180V-11