FABRICATION OF MIRROR SEGMENTS for the GSMT

Size: px

Start display at page:

Download "FABRICATION OF MIRROR SEGMENTS for the GSMT"

Rosalyn Melton
5 years ago
Views:

1 FABRICATION OF MIRROR SEGMENTS for the GSMT Segment Fabrication Workshop May 30, 2002

2 The USA Decadal Review In May 2000, the US astronomy decadal review committee recommended the construction of a 30-meter Giant Segmented Mirror Telescope (GSMT) as its highest-priority ground-based initiative In response, the has initiated several efforts including development of a point design to help identify the key technical issues. This is described in detail at:

3 Optical Design Optical design: Classical Cassegrain M1 diameter: 30 meters M1 focal ratio: f/1 M2 diameter: 2 meters M2 focal ratio: f/18.75 Aperture stop at M2

4 Radio Telescope Structural Design Lightweight steel truss structure M2 supported on tripod Elevation axis behind M1

5 Initial Point Design Structure Concept developed by Joe Antebi of Simpson Gumpertz & Heger Based on radio telescope Space frame truss Single counterweight

6 Initial Point Design Structure Plan View of Structure Pattern of segments Typical 'raft', 7 mirr m across Gemini Special raft - 6 places

7 Current Structural Concept

8 Segment Layout Segments grouped into rafts 7 segments per raft 9G 9B 9C 16C 16 types of rafts 91 rafts total 6G 9F 6B 6A 9A 9E 6C 9D 10G 10F 16B 10B 10A 16A 10C 10D 16D 13G 13F 13B 13A 13E 13C 13D 14B Ø30M 6F 6D 10E 11B 14G 14C 6E 7B 11G 11C 14A 3B 7G 7C 11A 14F 14D 3G 3C 7A 11F 11D 14E 15B 3A 7F 7D 11E 12B 15G 15C 3F 3D 7E 8B 12G 12C 15A 3E 4B 8G 8C 12A 15F 15D 4G 4C 8A 12F 12D 15E 4A 8F 8D 12E 4F 4D 8E 4E 5B 2B 5G 5C 2G 2C 5A 2A 5F 5D 2F 2D 5E 2E 1

Segment Properties Segment dimensions 1.15-m across flats -- 1.

9 Segment Properties Segment dimensions 1.15-m across flats m corner to corner 50 mm thickness Segment weight: 157 kg if Zerodur; 133 kg if ULE Number of segments: 618 Maximum departure from sphere 110 microns Comparable to Keck

10 Segment Supports Segment Supports Axial support: 18-point whiffletree FEA Gravity deflection 15 nm RMS Lateral support: 3 bipods -- line of action at mid-plane FEA Gravity deflection 2.2 nm RMS

11 Alternative Configurations 1. Lightweight structured segment supported on 3 points Material: SiC or Beryllium Advantages Lighter weight Simplified support Disadvantages Limited feasible size Fabrication Gravity sag Higher cost (?)

12 Alternative Configurations 2. Facesheet supported on graphite-epoxy truss Material: zero expansion glass or glass ceramic Truss connected to facesheet by displacement actuators Advantages Higher stiffness Active optics control of segment figure Disadvantages Residual thermal effects must be compensated by active optics

13 Fabrication Challenges Mechanical dimensions accurate to ~ 0.1 mm Bevel size ~ 0.5 mm Surface figure accuracy ~ 20 nm RMS Production of ~ 1000 segments in ~ 5 years Optical testing is challenging Absolute measurement of radius of curvature Figure accuracy Positional accuracy Large number of different: segment shapes orientations asphericities

14 We view segment fabrication as primarily a mass-production challenge. Cost Schedule Quality control

15 Fabrication Issues Our thinking on segment fabrication has been influenced by the work done for CELT: Terry S. Mast, Jerry E. Nelson, Gary E. Sommargren, Primary Mirror Segment Fabrication for CELT, SPIE 4003, Optical Design, Materials, Fabrication, and Maintenance, Munich, Germany, pp , 2000.

16 Bend-and-Polish Approach Warp segments and polish on CP machine Relax warping for optical testing Most efficient if segment can be cut to hex first Design of warping harness is tricky: Leave edges clear Low center of mass Clearance for weights on back of glass Avoid distortion on removal of harness

17 Bend-and-Polish Approach We considered using the mirror support whiffletree as a warping harness, with torque actuators at the pivots

18 GSMT segment at edge GSMT sample segment at the edge (r =15 m) P-V = 191 µm; RMS = 38 µm Astigmatism Coma 2 2 a r Z 4 = 112µ m 4R 3 3 a r Z 6 = 10µ m 2R 3

Active Force vs Residual error 1 2 1 1 1: No smoothing (residuals): P-V=12µm; RMS=2µm; Fmax=4014 N 3 2: less

19 Active Force vs Residual error : No smoothing (residuals): P-V=12µm; RMS=2µm; Fmax=4014 N 3 2: less smoothing (residuals): P-V=97µm; RMS=17µm; Fmax=504 N 3: more smoothing (residuals): P-V=169µm; RMS=33µm; Fmax=105N

20 Questions What processes are most cost-effective for finishing meter-sized off-axis aspheric segments? Computer controlled polishing (CCP)? Bend and polish? Ion milling? Replication? How much will 700 square meters of segments cost? What studies can best advance our understanding of segment fabrication methods and costs? How can we interest commercial optical finishers in developing low-cost segment fabrication techniques?

Design and Manufacture of 8.4 m Primary Mirror Segments and Supports for the GMT

Design and Manufacture of 8.4 m Primary Mirror Segments and Supports for the GMT Introduction The primary mirror for the Giant Magellan telescope is made up an 8.4 meter symmetric central segment surrounded