Assembly and Testing of the ESI Camera

A.I. Sheinisa, B. Sutinb, H. Eppsc, J.A. Schierd, D. Hilyarda, J. Lewisa

 

a UCO/Lick Observatory, University of California, Santa Cruz, MS 654, Santa Cruz, CA 95064

b Carnegie Observatories, 813 Santa Barbara Street, Pasadena, CA 91101

c UCO/Lick Observatory, University of California, Santa Cruz, Kerr Hall, Santa Cruz, CA 95064

d J. Alan Schier Co , 3130 Foothill Blvd., Unit #1, La Crescenta, CA 91214

 

The Echellette Spectrograph and Imager (ESI), currently being delivered for use at the Cassegrain focus of the Keck II telescope employs an all-spherical, 308 mm focal length f/1.07 (underfilled) Epps camera. The camera consists of 10 lens elements in 5 groups: an oil-coupled doublet; a singlet, an oil-coupled triplet; a grease-coupled triplet; and a field flattener, which also serves as the vacuum-dewar window. A sensitivity analysis suggested that mechanical manufacturing tolerances of order +/- 25 microns were appropriate. In this paper we discuss the sensitivity analysis, the assembly and the testing of this camera.

Keywords: spectrograph, imager, echellette, Keck, camera, astronomy, prism.

 

1. Introduction:

 

The Echellette Spectrograph and Imager (ESI) is a multipurpose instrument currently being delivered by the Instrument Development Laboratory of Lick Observatory for use at the Cassegrain focus of the Keck II telescope. The P.I. for the instrument is Joe Miller, director of Lick Observatory. ESI is scheduled for first light on August 29, 1999. ESI is a medium resolution spectrograph, with a fixed 175.6 line/mm echellette grating from Milton Roy (now Spectronics, Rochester, NY 14625) and two large (25 kg) cross-dispersion prisms made of BSL7Y glass from Ohara Corp., (San Clemente, CA). The prisms were polished by Zygo (Middlefield, CT) and coated with multi-layer dielectric anti-reflection coatings (AR) by Coherent Inc., (Auburn, CA). ESI also contains a reflective collimator which is an off-axis segment of an on-axis ellipsoid, figured by TORC, (Tucson, AZ) and coated by Newport Thin Films Laboratory, (Chico, CA). ESI currently has three scientific modes available: a medium resolution Echellete mode; a low resolution prismatic mode; and an imaging mode . ESI is described and referenced in detail in several previous papers 1,2,3,4,5.

The spectrograph and camera are designed to accommodate the (0.39 to 1.10)-micron passband without refocus. The spectrum is detected by a single flat (2048 by 4096 by 15-micron) thinned, back-illuminated, MIT-Lincoln Labs CCD. The entire spectrum is recorded in a single exposure for both the echellette mode and the low dispersion prismatic (LDP) mode. The echellette mode accommodates a 20 arcsec-long slit and has an average resolution (l /D l ) of 26,087 per pixel. The LDP mode takes multiple spectra of as many as 50 objects or along a single 8 arcmin long slit at a per-pixel resolution (l /D l ) of 4839 (at 0.39 microns) to 1052 (at 0.80 microns). The imaging mode covers a (2 by 8)-arcmin field of view at 0.153 arcsec per pixel.

The operating temperature range for both the spectrograph and the camera is û4 C to +6 C. The survivability range is û20 C to +30 C.

The camera optical design, shown in Figure 1, was conjured by one of the authors (H.E.). It consists of ten lens elements in five lens groups. The camera has an effective focal length of 308 mm. It has an entrance aperture diameter of 287 mm and a final plate scale of 97.7 microns/arcsec on the sky. The collimated beam diameter is approximately 160 mm. The cameraÆs effective f/ratio in imaging mode is thus f/1.93 and slightly faster in spectroscopic modes due to anamorphism.

The camera design is all-spherical and it includes two large CaF2 lenses. Group #1 is a doublet, group #2 is a CaF2 singlet, groups #3 and #4 are triplets while group #5 is the field flattener/dewar window. The elements in groups #1 and #3 are optically coupled with a fluid (Cargille laser liquid Type 5610 n(D)=1.5000) to minimize internal reflections. The elements in group #4 are greased together with Dow Corning Q2-3067 optical couplant. The six larger elements were fabricated by TORC and the 4 smaller elements were fabricated by Cosmo Optics Inc., (Middletown, NY). Broad passband AR coatings were applied by Coherent.

 

2. Collimated System Performance:

The ESI optical design is complicated by the fact that a wide variety of pupil anamorphic factors and effective entrance pupil distances are presented to the cameraÆs entrance aperture by reason of the three operating modes. In practice, the camera design was slightly compromised in the imaging and LDP modes so as to favor the echellette mode. Nevertheless, the echellette mode remains the most severe test of system image quality. The system performance is described in considerable detail by Sutin8, in which he used the pre-construction ESI camera Run No. 2990 (02/27/96) by Epps9.

The system performance was recalculated using the final as-built camera Run No. 102297AC. Orders 6 in the infrared through 15 in the ultraviolet were raytraced at five wavelengths uniformly spaced over each free-spectral range without refocus. The rms spot-size diameter (Drms) was calculated for each image. Averaging these image diameters over all wavelengths and all orders the Ave(rms) = 19.0 +/- 3.3 microns. The corresponding 80% encircled ray diameter average is Ave(80%) = 22.5 +/- 4.2 microns, while Ave(90%) = 28.4 +/- 5.6 microns.

3. Sensitivity Analysis:

A detailed computer-based sensitivity analysis was divined by one of the authors (B.S.). Individual lens elements were perturbed axially, radially, and in tip/tilt. The perturbations were 25 microns for axial (Z) and radial (Y) movements. The tip/tilt (R) perturbation angles about each lens element center were the Arctan(25 microns / element clear-diameter (cd)). The perturbed camera model was ray-traced within the ESI spectrograph numerical model in echellette mode at 50 different wavelengths, covering all 10 orders, with 1000 rays in each monochromatic spot diagram. The perturbed system was then compared to the unperturbed system and the worst centroid location change and worst increase in the rms spot-size diameter (Drms) were noted. For the first set of results, no re-focus or active centroiding was done. These calculations correspond to the worst-case scenario. For the second set of results, the median lateral image centroid shifts were subtracted, and a focus correction was estimated and applied. This case corresponds approximately to an active collimator scenario.

In addition, a similar analysis was performed for the lens groups. Sensitivities were determined by moving entire groups and calculating the resulting effects on centroid location and Drms. The following Table summarizes this analysis:

 

 

 

The columns are:

  1. lens element number, starting at the first element
  2. Degree of Freedom

3) perturbation size (always 25 m or 25/cd mrad)

4) and 6) maximum centroid location change, in m .

5) and 7) maximum increase in Drms, in m .

ESI Camera Sensitivities

Brian Sutin (11/96)

no active corrections

active corrections

lens #

Degree of

Freedom

Distance

(m ) or

mrad

Image motion

(m )

Spot size increase

(m )

Image motion

(m )

Spot size increase

(m )

1

Y displacement

25

7.699

0.192

0.185

0.196

1

Z displacement

25

0.219

0.435

0.241

0.035

1

Tip/Tilt

25/cd

10.33

0.509

0.37

0.542

2

Y displacement

25

4.683

0.443

0.185

0.517

2

Z displacement

25

0.087

0.315

0.065

0.01

2

Tip/Tilt

25/cd

12.179

0.132

0.287

0.139

3

Y displacement

25

13.182

1.047

0.722

1

3

Z displacement

25

1.837

3.389

1.721

0.93

3

Tip/Tilt

25/cd

3.678

0.095

0.072

0.052

4

Y displacement

25

13.363

0.277

0.398

0.257

4

Z displacement

25

3.762

4.021

3.672

0.4

4

Tip/Tilt

25/cd

8.747

0.103

0.292

0.117

5

Y displacement

25

1.769

0.221

0.246

0.222

5

Z displacement

25

0.754

0.822

0.736

0.142

5

Tip/Tilt

25/cd

11.166

0.29

0.315

0.299

6

Y displacement

25

4.871

0.267

0.22

0.3

6

Z displacement

25

0.768

0.057

0.773

0.019

6

Tip/Tilt

25/cd

10.12

0.325

0.298

0.268

7

Y displacement

25

4.918

0.02

0.631

0

7

Z displacement

25

0.957

0.326

0.965

0.333

7

Tip/Tilt

25/cd

5.37

0.27

0.56

0.219

8

Y displacement

25

2.411

0.028

0.175

0.014

8

Z displacement

25

2.104

1.276

2.054

0.293

8

Tip/Tilt

25/cd

16.842

0.389

0.246

0.375

9

Y displacement

25

3.298

0.387

0.323

0.407

9

Z displacement

25

3.348

0.979

3.372

0.179

9

Tip/Tilt

25/cd

4.076

0.187

0.742

0.116

10

Y displacement

25

1.016

0.303

0.215

0.291

10

Z displacement

25

2.668

0.963

2.645

0.369

10

Tip/Tilt

25/cd

1.351

0.335

0.461

0.321

group 1

Y displacement

25

3.22

0.247

0.128

0.219

group 1

Z displacement

25

0.275

0.119

0.276

0.025

group 1

Tip/Tilt

25/cd

1.713

0.671

0.219

0.745

group 3

Y displacement

25

6.953

0.768

0.164

0.792

group 3

Z displacement

25

2.266

3.177

2.158

0.163

group 3

Tip/Tilt

25/cd

10.714

0.43

0.285

0.406

group 4

Y displacement

25

3.981

0.312

0.403

0.262

group 4

Z displacement

25

0.459

2.26

0.558

0.311

group 4

Tip/Tilt

25/cd

11.988

0.273

0.051

0.276

 

These sensitivities were used to estimate manufacturing and assembly tolerances for the mechanical components such that the expected summation of image errors (added in quadrature) was comparable to the residual aberrations in the optical design. In addition, the sensitivity data also allowed us to set a lens motion tolerance in terms of the final image motion allowable for different gravity orientations.

 

 

 

4. Camera Optomechanical Design

The camera mechanical system (CMS) for ESI was designed by one of the authors (J.A.S.), (J. Alan Schier Co., La Crescenta, CA 91214) and fabricated by Danco Machine, DPMS Inc (Santa Clara, CA). The CMS is shown in Figure 2. It consists of four cells supported within a single large barrel. The mass of the entire camera assembly, consisting of the CMS and the lens elements is approximately 125 kg. The individual cells with their respective lens elements have masses ranging from 5 to 30 kg. The cells locate radially against locating surfaces (lands) on the barrel inner wall and axially against athermalizing spacers. Each cell consists of an aluminum housing, radial athermalizing spacers and a compressive preload spring. Each of the two oil-coupled cells also contains an oil sealing reservoir system. The lens element spacings within the oil-coupled cells are maintained by 0.10-mm Mylar spacers, placed between the lens elements at their edges. We now describe the four major features of the optomechanical design:

Axial Athermalization:

In principle, athermalizaton would require that the effective system magnification as well as the focal location relative to the CCD are invariant over the operating temperature range. As part of the sensitivity analysis (B.S.) calculated a linear relationship between the two intergroup spacings, such that the optical impact of small perturbations (all axial distances, changes in curvature and changes of refractive index) was minimized with regard to magnification and focal location. In practice he found that the compensation was imperfect . It was decided to emphasize focal position over image scale change.

From these considerations, he determined the effective coefficients of thermal expansion (CTE) between the two individual groups so as to provide the required spacing changes with temperature. These effective CTEÆs were achieved by using stacks of 3 materials, (aluminum, Delrin II and polyurethane ) between cells 2 and 3 and between cells 3 and 4. A single material (aluminum) was adopted between cells 1 and 2 and between cell 4 and the dewar.

Actual material CTEÆs had to be measured in the laboratory and in some cases they differed substantially from the vendorsÆ catalog values. In addition, simple material compression had to be considered in the material choice as the weight of the cells plus compression spring contributed to a possible compressive force of up to 130 Kg on the spacers.

Radial Athermalization:

The purpose of the radial athermalization is to maintain zero internal stress in all the optical elements over the entire survivability temperature range while maintaining an axial centration tolerance less than that prescribed by the considerations in Section 3 over the operating temperature range.

To achieve this goal, a set of Delrin II pads was used to match material expansion with temperature, due to the difference in the CTEÆs of the glass lenses and the aluminum housings. This concept is similar to that used in the LRIS camera7 on Keck I. The pad thickness was chosen based on the CTEÆs of the different glasses. With a diameter clearance of 0.05 mm between the glass and Delrin II pads, all cells have zero thermally induced stresses over the previously mentioned survivability range. There is no upper bound as the clearance increases with increasing temperature.

An additional factor of great concern is the effect of transient temperature changes. If the ambient temperature cools too rapidly, a thermally induced interference can occur within the cells. This results from the fact that the aluminum housings have thermal time constants of approximately 15 minutes while the lenses have thermal time constants as long as 120 minutes. At a temperature difference of 14 C an interference occurs in cell #2. The other cells experience this at greater temperature differences. With this in mind, we have set 5 C per hour as the maximum temperature change rate to be experienced by the camera under any circumstance.

Cell #2 Transverse Adjustment:

It can be seen in Table 1 that cell #2 has the highest sensitivity to radial translation. One desire that came out of the sensitivity analysis was to have cell #2 adjustable in translation after assembly. This was desired for two reasons: 1) We felt we couldnÆt achieve the desired lateral location for cell #2 passively, given the current design concept and 2) It was noted that translational adjustment of cell #2 could compensate for other fabrication and assembly errors in other parts of the CMS.

Figure 3 shows cell #2. A 0.5 mm passage has been cut completely through the aluminum housing except for three transverse flexures. This cut effectively creates an inner and outer housing. A preload spring is mounted in the outer housing along with two adjusting screw assemblies. This allows the inner housing to be adjusted transversely relative to the outer housing. In this way transverse adjustment of cell #2 is achieved.

Optical Coupling Fluid Control:

Cells #1 and #3 contain an optical coupling fluid (previously described) to minimize internal reflections between elements within the cell. An o-ring seal is provided at the glass-aluminum interface. The o-ring is compressed at the minimum nominal-dynamic-seal specification (8%). The o-ring is compressed such that the optic locates directly against the aluminum land and the o-ring carries most of the load, while providing a seal.

A polyurethane bladder is used to accommodate volumetric changes of the coupling fluid within each cell. Possible reactivity of the optical coupling fluid with polyurethane and other substances was tested (Hilyard6, et al., this Conference). They found no reactivity with the polyurethane bladder material.

 

5. Integration and Alignment Procedure

 

The camera integration procedure was developed interactively with the aid of a set of simulated lens elements fabricated from aluminum. It was practiced several times with the simulated lens elements until we were comfortable with the amount of manual handling required and the risks involved. With the exception of the translational location of cell #2, the cameraÆs optical alignment is achieved by close-tolerance fabrication of the CMS.

The radial locating surfaces within the cells, i.e., the inner radius of the Delrin II radial athermalizing spacers, as shown in Figure 3, has been final machined to match the as-measured isothermal optic diameters plus a 0.025 mm radial clearance. This was done after the spacers were bolted into the aluminum housings in order to increase the centration accuracy of the hole relative to the outer housing diameter.

The cells were then assembled on a precision rotary table. Both the lens elements and housings were aligned to the rotary table by minimizing their horizontal and vertical run-out. Next the optics were held rigidly in place while the housings were jacked into position. Optics retainers with pre-load springs were then installed into each assembly. The rest of the mechanical components were then installed onto each cell, including the oil handling systems, for doublet and large triplet.

The oil was added to the cells via a gravity feed system. The oil had been previously placed into an evacuated bell jar to remove any dissolved gasses. If bubbles should appear at a later time, it may prove necessary to refill the cells under vacuum.

After the cells were assembled, cell-land locations were measured relative to the lens vertices. Only then were the final lengths of the axial spacers cut. This was done to remove as much axial assembly error stack-up as possible.

Axially, cell #1 locates against the front of the barrel via an internal land. All other cells locate against the back of the barrel through the other cells and associated spacers. A compliant spacer with a pre-load spring provides the necessary force between cell #1 and cells 2, 3 and 4 along with their spacers, which make up a stack whose axial thermal compensation is described in Section 4.

For cells #1, #2, and #3 a hand operated hoist is used to lower the cells into the barrel one at a time. Cell #4 is jacked into place from below. Measurements are made of the "stack height" as the cells are installed to confirm that the cells are seated properly on the spacers.

As mentioned previously, lands within the barrel determine the lateral location of the cells. These barrel lands have been final machined in a single lathe operation such that they are concentric to +/- 0.025 mm. Finally, they were honed individually to match the as-measured isothermal cell diameters. The resulting radial clearance per cell was left at .025 mm, for ease assembly. Radial shims were added afterward to take up these gaps.

6. Camera Bench Testing and Performance Analysis

The ESI camera was designed to operate under the ambient Keck observatory conditions. However, the integration and testing were carried out at the Instrument Development Laboratory of Lick Observatory in Santa Cruz, CA, which differs substantially in ambient temperature, pressure and humidity. Also, some of the optical tests were performed without oil in the multiplets. These effects were considered and accounted for in all calculations related to the integration and testing. However the details of these corrections are beyond the scope of the present paper.

After Laboratory assembly of the camera, several optical performance tests were carried out. The results of these tests were then compared to the predictions of the lens design and analysis. We now describe two different tests.

Interferometric test:

A double pass interferogram was taken of the camera by sending an f/0.7 converging .633-micron beam from a Zygo Mark IV interferometer into the image plane, through the camera and returned off of a flat reference mirror. The results of this test were compared to synthetic interferogram, calculated with the Zemax lens design program (Focus Software Inc., Tucson, AZ) . Figure 4A shows the synthesized interferogram produced by Zemax. Figure 4B shows the actual interferogram. Note that the central obscuration in Figure 4B is artifactual.

We were pleased to see that the interferometric test matched the prediction of the optical design to high precision.

 

Point Spread function (PSF) and Back Focal Distance (BFD) measurements:

A polychromatic light source was defined with a broad passband optical filter (Wratten 54). This light source illuminated a 5-micron diameter pinhole at the focus of a 1219-mm focal length laboratory telescope, producing a 152-mm diameter collimated beam. This illuminated the cameraÆs entrance aperture at zero field angle. The image formed by the camera was relayed to a Cohu CCD via a 20X microscope. This image energy distribution was digitized with a Matrox frame grabber board. The digital information was converted to a PSF using the Information Data Language (IDL). Figure 5A shows a theoretical PSF produced by Zemax while Figure 5B shows the actual measured PSF. While the sizes are comparable, the shapes differ slightly. The differences can be attributed to known residual aberrations in the laboratory telescope.

The FWHM of the polychromatic spot was determined from the PSF to be 10.6 microns. This is comparable to the limitation of the design. The BFL was also measured and was found to match, to within 12 microns, that predicted by the design on the first assembly prior to oil fill. However, the BFL measure was too large by 0.2 mm at the final assembly measurement. The most likely candidate for this error has been determined to be the reduction of intra-cell distances in cell #4 due to the leakage of some of the coupling grease in that cell. This issue will be addressed during the commissioning at Keck in August 1999 by regreasing the last group before reassembly.

7. Conclusion:

We have described the optomechanical design, assembly and testing of a 10-lens element, all spherical f/1.07 Epps camera. The camera is part of the ESI spectrograph, which is scheduled for first light on August 29, 1999. The results of our in-house camera-performance testing matched the predictions of the optical design to high precision. Further proof of the cameraÆs performance under actual observatory conditions will be seen at the Keck telescope in Hawaii.

 

8. Acknowledgments

The authors wish to thank Carol Osborne and Mary Poteete for their help with the design drafting and many of the Figures for this paper. We would like to thank George Laopodis for his assistance with the integration of the optics. Also, we would like to thank Joe Miller, Jerry Nelson, Terry Mast, Matt Radovan and Jack Osborne for their many helpful comments.

9. Bibliography

 

  1. Epps, H., and Miller, J., "Echellette Spectrograph and Imager for Keck Observatory," Proc. SPIE 3355, pp. 48-58, March 1998
  2. Sutin,B., "What an optical designer can do for you AFTER you get the design," Proc. SPIE 3355, pp. 134-143, March 1998
  3. Sheinis, A.I., et al., " Large prism mounting to minimize rotation in Cassegrain instruments," Proc. SPIE 3355, pp. 59-69, March 1998
  4. Radovan, M. V. ,et al., "Design of a collimator support to provde flexure control on Cassegrain spectrographs," Proc. SPIE 3355, pp. 155-163, March 1998
  5. Bigelow,B.C., and Nelson,J.E., "Determinate space-frame structure for the Keck II echellette spectrograph and imager (ESI)," Proc. SPIE 3355, pp. 164-174, March 1998
  6. Hilyard, D., "Chemical Reactivity Testing of Optical Fluids and Materials in the DEIMOS Spectrographic Camera for the Keck II Telescope," (these Proceedings)
  7. Oke, J.B. et al., "The Keck Low-Resolution Imaging Spectrometer," PASP 107, pp. 375-385, April 1995
  8. Sutin,B., "ESI: a new spectrograph for the Keck II telescope," Proc. SPIE 2871, pp. 1116-1125, June 1996
  9. Epps,H., "Development of large high-performance lenses for astronomical spectrographs," Proc. SPIE 3355, pp. 111-128, March 1998