Designs of null test optics for 8.4-m, ƒ/1.1 paraboloidal mirrors Jim Burge Several null lenses are considered for measuring the primary mirrors for UA’s.

Designs of null test optics for 8.4-m, ƒ/1.1 paraboloidal mirrors

Jim Burge

Several null lenses are considered for measuring the primary mirrors for UA’s Large Binocular Telescope

• Infrared null lens using a diamond-turned asphere • Giant refractive Offner-type null lens• Gregorian variation of Offner reflective design• Null lens using a binary computer-generated hologram

Optical Sciences Center and Steward ObservatoryUniversity of Arizona

U of A is making the world’s steepest large primary mirrors

1700 1750 1800 1850 1900 1950 2000 2050year

0.01

0.10

1.00

10.00

100.00

1000.00

10000.00

p-v

asph

ere

(µm

)

LBT

Hale

MMT, Magellan

Herschel

Primary mirror measurements are difficult because

of the large surface departure from spherical

-5000 -4000 -3000 -2000 -1000 0 1000 2000 3000 4000 5000position on mirror(mm)

0

400

800

1200

1600

dep

artu

re f

rom

sp

her

e (µ

m)

LBT 8.4-m

M M T 6.5-m

VA TT 1.8-m G em in i/VLT8-m

AF 3.5-m

IR null lens using diamond turned asphere

• Uses 10.6 µm light from CO2 laser• Similar to successful design used for 6.5-m mirrors, re-uses Ge lens• DT asphere gives perfect wavefront and excellent imaging• Ultimate accuracy is less important for IR than visible• Calibrate with Computer Generated Hologram to 0.1 µm rms

1 .3 m e te rs

5 0 m mZ nS ew /a sp he re

2 0 0 m m d ia m G ep lano -co n ve x len s 1 7 0 m m d ia m

Z nS e le n s

P a rax ia l fo cus(1 9 .2 m e te rs to

p rim ary m irro r)

T w y m an n -G reenin te rfe ro m e te r

Previous IR null lens for 6.5-m ƒ/1.25

CGH measurement of this null lens shows 0.02 rms error

Includes 0.007 rms low order spherical aberration

Null lens evolution

Paraxial focus

3.5-m ƒ/1.75

6.5-m ƒ/1.25

8.4-m ƒ/1.14

Offner-type refractive null

• Similar to previous designs• Design gives excellent correction• Limited by glass quality in large lens• Manufacture of large, fast convex surface is difficult

-.02

0.003 rms

2.1 meters

390 mm diameter BK7 lens90 mm thickR/0.74 convex sphere Paraxial

focusinterferometerfocus

Gregorian version of Offner reflective null

• Uses mirrors to solve index problem• Gives excellent performance• Has been analyzed in detail using structure functions• Difficult opto-mechanical design

SPHERICAL PRIMARY750 mm diam

MANGIN SECONDARY50 mm diam

TWO FIELD LENSES90 mm diam

2 METER TOTAL LENGTH

TO PRIMARY MIRROR

COLLIMATED INTERFEROMETER

-.02

0.002 rms

CGH null lens

• Uses 2 CGH’s– Illumination CGH controls slope for both reference and test wavefronts– Reference CGH creates reference wavefront

• Compact design, • Can phase shift by pushing reference CGH with PZTs• Needs more careful study

Reference CGH (reflects reference wavefront, transmits test wavefront)150 mm diam

IlluminationCGH(etched)

Point source/image

180 mm diam38 mm thick lens

200 mm diamplano substrates

1.1 meters

CGH creates reference wavefront

19 m to primary

Point source/image

Illumination CGH CGH to create reference wavefront

Reference beam

-1 order Littrow diffraction

Test Beam

0 order twice through CGH

CGH design

• Requires ~12,000 rings, each accurately placed• This CGH is easily within modern fabrication capabilities• CGH fabrication errors will contribute 3 nm rms to surface error

0 20 40 60 80radial position in mm

0

2

4

6

8

10

12

14

16

18

20

rin

g s

pac

ing

in µ

m

Candidate null corrector designs for 8.4-m ƒ/1.14 primary mirrors

Null lens certification with CGH

• 58,000 rings• 200 mm diameter• Measures conic

constant to accuracy of <0.0001

-120 -80 -40 0 40 80 120radial position on hologram (m m)

0

20000

40000

60000

CG

H O

PD

in w

aves

LBT

M M T

LBT CGH is easier than previous

0 20 40 60 80 100 120radial position on hologram (mm)

0

2

4

6

8

10

rin

g s

pac

ing

(µm

)

LBT

M M T

CGH for 8.4-m f/1.1 at 632.8 nm has larger features than previous, successful CGH for 6.5-m f/1.25 at 530.7 nm

CGH fabrication verified

• f/1.14 holograms were manufactured by group from Russian Academy of Sciences

• Wavefronts were measured interferometrically• Figure accuracy of 5 nm rms is typical

Conclusion

• The jury is still out on the type of null corrector for LBT• A different important issue still needs to decided --

– Holographic certification has been extremely successful – The holograms are intrinsically more accurate than the null correctors– What about aligning the null corrector based on the hologram?– This would save a lot of money and time– We could use a second, independently made hologram as verification