We can envisage a narrowband tunable filter for the Cassegrain or Nasmyth focus of a 30m-class telescope.  Inevitably, cost and resource considerations will require a tunable filter device to share the focal reducer space of a low- to medium resolution (R=300-3000) grating spectrograph with a pupil size of 300 mm or more.

Particularly attractive is the capability of 30m-class telescopes to access very faint, high-z objects which would require infrared capability for the tunable filter. We take as our base design a cryogenic Fabry-Perot tunable filter comprised of two plates with either dielectric or metal coatings, with an airgap controlled by servo-actuators. 
 
 

1. Taurus Tunable Filter (TTF) image of the Orion nebula taken in 0.6” seeing at the AAT. This image is produced from three emission lines: H-alpha (blue), [NII] (green), [SII] (red).

3.  Principal Goals

[Describe the key aspects and requirements of the technology development in quantitative terms, for example, in terms of numbers of actuators, electrons of read noise, unit cost targets, etc.  These goals should be for the entire development program, not just the first three years.  Provide an overview of the schedule and deliverables for the entire program (the detailed plan for the DDP is in the sections below)] 
 

For our proof of concept, we adopt the optical design of the proposed IR imager (web link to Sec. 4.7.4 GSMT handbook). In this optical design, a pupil size of 485mm gives an imager field of view of 1.44 arcmin. In practice, it is accepted that this may be limited to ~300mm (1 arcmin) by the availability of CaF₂ optics of this size. We take 300mm as our design goal for a tunable filter which would have wide application on all large aperture telescopes.

The world leaders in Fabry-Perot based tunable filters are ICOS (formerly Queensgate Ltd). Their existing designs cannot be efficiently utilized by large telescopes (> 6m diameter) since these devices are bulky, have small working apertures (150mm or less), and are optimized for optical applications. These devices operate poorly in cryogenic environments. Cryogenic devices already under development have small apertures, e.g. a proposed design for the mid-infrared JWST DULCE imager is based on a 40mm aperture.

Our primary goals are to develop technologies leading to the production of a tunable Fabry-Perot etalon with a usable aperture approaching or exceeding 300mm, and suitable for cryogenic operation.

Specific areas requiring technological development and prototyping are:

• larger unsupported aperture imposes flatness requirements difficult to meet from both manufacturing perspective and from plate stiffness implications.
• Greater required plate stiffness leads to heavier plates and consequent greater lateral displacement at varying alt-az.
• flatness requirements apply to coatings as well as substrate flatness.
• the likely need to operate the etalon with more actuators and sensors (the now-dated Queensgate control system is limited to three actuators).
• cryogenic operation imposes additional constraints, particularly on the servo-actuators, however certain new technologies are likely to enable the required actuator material specifications.

Proposed system.  We propose to use current technologies to develop a servo-actuator comprised of piezo actuator and capacitive plate gap sensor, with a digital control system.The advantage of this modular approach to position control is that the digital control system allows an arbitrary number of actuator/sensor pairs to be combined with a dynamic mechanical model of the etalon, taking into account various other system and environmental parameters such as the direction of the gravity vector, and temperature fluctuations and gradients in the device.

Extension of the device aperture to ~300mm will be with the use of four servo-actuators – three peripheral and one central. The device is to be used in a pupil image, taking advantage of the central obstruction to accommodate the central actuator. This effectively limits the applications to Cassegrain or Nasmyth-based instruments, however this requirement is viewed as consistent with present Large Telescope concept designs.

Possible use of sapphire plates for IR application gives good transmission over the wavelength range of interest along with greater stiffness that leads to thinner and lighter plates.  In the proposed configuration, light only passes through the device at small angles to normal and so the birefringence of sapphire should not cause a significant problem.

The central (and perhaps other) actuators can have cross-electrodes after the manner of the Echidna spine actuators, providing the additional capability for application of force in arbitrary directions parallel to the etalon plane.  In combination with knowledge of the gravity vector with respect to the device, this capability may be used to maintain the alignment of the capacitive sensors and avoid capacitor misalignment errors.

A central actuator reduces plate flatness requirements to a local requirement applicable only over ~150mm, which is much less demanding and well within present manufacturing capabilities. 
 

2. Plan view of conceptual design of large-aperture tunable filter based on a Fabry-Perot etalon employing four piezo actuators with capacitive plate nanometry. The electrical feeds to the central components will be masked by the obscuration of the telescope’s secondary mirror support structure, preserved in a pupil image.

We investigate metal coatings as these do not suffer from reflectance and phase nonlinearity at small spacings, when the airgap approaches the coating thicknesses (Jones and Bland-Hawthorn 1998).

Our investigations of nanopositioning feedback technologies will focus on capacitive plate sensor techniques, but we extend consideration to the possible use of out-of-band laser sensing of plate separation (this could potentially use light for metrology with a wavelength outside operational detector sensitivity – e.g. short wavelength visible for sensing, in an IR imager). 
 

4.  Key Tasks for First Three Years

[Prepare a work breakdown structure (WBS) listing the main steps in the first three years of the development effort, such as: Select vendors;  Prepare feasibility study report; Fabricate sub-scale prototypes; etc.  Write a short paragraph describing each task.  Describe how these tasks support the long-term goals, and indicate how imperative it is to accomplish these tasks in the next three years.] 
 

Tunable Filter

Work Package 1

Specify requirements, confirm project scope.

Work package 2

Research available technology (filter technology, nanopositioning technology, plate construction materials and coating technology.

Work package 3

Negotiate with industrial partners – assess viability of technology solutions and cost benefit for define work in Work Package 2.

Decision Point.

Decision to proceed with further development or to rule out large aperture tunable FP etalon as a viable tunable filter. Possible switch to alternative technologies (plate materials, geometry and coatings, actuators and position feedback sensors to be considered). Decisions here based on considered expert assessment resulting from negotiation with partners.

Work package 4

Develop concept design for tunable filter, including basic filter parameters, FEA modeling and system performance modeling.

Decision Point.

Decision to proceed with component prototyping phase, based on system modeling.  Possible switch to alternate technologies or design.

Work package 5

Prototype high risk components: cryo servo actuators, materials and coatings, control systems.

Work package 6

Test and characterize prototypes produced in work package 5.

Decision Point.

Decision to proceed with concept demonstration prototype instrument, based on component prototype results.  Possible modifications to design.

Work package 7

Develop prototype filter suitable for ‘on sky’ measurements.

Work package 8

Characterise filter performance using ‘on sky’ tests.

Work package 9

Report on findings giving recommendations for a full scale tunable filter suitable for ELT

5.  Summary of Deliverables

[Indicate the main deliverables in the first three years.]

Subject to project termination at the defined decision points, deliverables for this technology development package are as follows:

1. A report detailing the present state of the art in tunable filter technology and presenting the details of research into currently available technologies suitable for development of a large aperture device suitable for cryogenic operation.
2. A document detailing a concept design for a large aperture tunable etalon, including basic filter parameters, proposed geometry, coatings, positioning and sensing technologies, and both FEA and optical modelling.
3. Prototypes of major system components (at a minimum, servo-actuators suitable for cryogenic operation and digital control, and a prototype digital control system).
4. Prototype etalon and support instrument suitable for on-sky testing.
5. Report detailing prototype performance and recommendations for proceeding with a full scale device.

6.  Schedule

[Show a schedule (in a MicroSoft Project Gant chart) for the activities listed under Key Tasks, including an indication of dependencies.  The schedule should include decision points indicated as milestones.  Assume the DDP will run three years starting October 1 2003, but simply indicate the time in the schedule as "Year 1, Year 2, etc.")]

The complete Master Schedule is available upon request.

7.  Decision Points

[Describe the main decision points that will occur in the first three years.  These could be the points at which you decide which of several competing technologies to pursue, or they could be the points at which you decide whether to continue or cancel a part of the program.  If possible, describe the criteria that will be used to make the decisions, and the anticipated review processes.  Include a flow chart illustrating the logic of the decision-making process, if appropriate.]

Decision points are shown in Work Packages and Schedule.

8.  Cost Estimate

[Estimate the cost for each task, or group of tasks, in the WBS.  This should include the cost of: purchased equipment and supplies; subcontracts; travel; purchased services; software; etc.  Obtain cost estimates from potential vendors or recognized experts, to the extent possible.  The cost of the project labor required to plan the tasks, write specifications, select vendors, supervise contracts, etc., should also be included.  Please estimate the number of man-months required in each of the following staff categories:

Senior scientist

Associate scientist

Group manager

Senior engineer

Associate engineer

Administrative assistant

(As we pull the entire proposal together, we will agree on labor costs for each category, and convert the man-months to dollars.)]

Full details are included within the Master Schedule (available upon request).

9.  Backup Plan

[Describe the backup plan in case the technology development is not as successful as hoped.]

Backup plan is included under Section 2.

Appendix A:         Industrial partners

ICOS Optics Design, Kent, UK: Formerly Queensgate, ICOS produced the Taurus Tunable Filter used on the AAT and WHT. This device is a clear model for the proposed device.

Barr Associates, Westford, MA, USA: Barr Associates are amongst the world leaders in multi-layer coating technologies and produce the highest quality astronomical filters.

CSIRO, Lindfield (National Measurement Laboratory), Australia: The world-class capabilities of the CSIRO laboratories already produce the highest quality surface finishes, for example in LIGO optics which have some of the most stringent specifications, and in tunable filters based on LiNO₃ solid etalons. They are a logical choice for collaboration on tunable filter development.

Appendix B:       Tunable Filter Plate Specifications

Bland-Hawthorn et al. (2001) find that the spectroscopic resolution of this type of Fabry-Perot etalon derives from an effective finesse N_E, largely determined by the coating reflectivity of the two plates, but degraded by non-uniformities in plate and coating flatness (N_D for small scale non-uniformities and N_LD for large scale distortions or bowing of the plates), and by the beam aperture finesse N_A. Addition of these effects is in quadrature, as shown in equation 1.

In order to minimize the impact of coating/plate defects, Bland-Hawthorn (1995, Fig. 1) finds that N_E ≈ N_R ≈ 40.

The aperture finesse is a degradation resulting from the range of angles over which the imaged light passes through the filter, and derives from the speed of the imaging system. For light entering the filter from a solid angle Ω (focal speed F), the filtered output of order m has an aperture finesse given by equation 2

Based on this relation, Bland-Hawthorn et al. (2001) show that beams slower than f/15 (F=15) are suitable for filters operating near N_E ≈ 40. This is entirely consistent with operation of the proposed tunable filter within the MCAO-fed near-IR imager conceived in section 4.7.4 of the GSMT Book.

Similarly, in order to allow operation at the required finesse of 40, the flatness finesse should approach  or exceed double this figure. This implies a flatness criterion near λ/100, and so this is taken as our baseline flatness specification, as well as our parallelism and bowing criteria. For NIR operation at 1 to 2 µm, this implies flatness and parallelism requirements across the entire aperture of the device in the order of 10 nm.