Instrument Architecture

The CHIME instrument is built upon the concept of a pushbroom imaging spectrometer. The instrument provides ~130 km of swath at 30m GSD employing spatial swath-splitting: a single wide field Three Mirror Anastigmat (TMA) telescope with a folding mirror feeds three independent identical full wavelength range (400-2500nm) ‘VISIR’ grating spectrometer back ends in a staggered slit configuration with overlap. The de-magnifying optical design allows the telescope to operate at F#/5.5, while the spectrographs are built in a more compact arrangement at F#/3.3. The Offner-type spectrometers employ dual-blazed gratings delivering 10nm SSI. Chapter 2 discusses the telescope and spectrometer optics in more detail.

The instrument is built around a central torus-like hollow CFRP frame structure that provides the necessary optomechanical stiffness in a single element. The telescope mirrors and sub-structures mount directly to the central torus as does the AOCS equipment. At the focal plane of the telescope, the three identical spectrometer units are mounted on a common spectrometer support structure that connects to the central optical frame torus from below to keep the CoG close to the platform. The structure of the instrument is further discussed in Chapter 3.

All units, except the three MCT detectors, work at ambient temperature. The three cold (~170K) detection back ends are built around the Teledyne CHROMA-D MCT 3k detector as a baseline. It features 3072 pixel in spectral and 512pixel in spatial direction. The pixel are binned 2x2 to decrease the NEDL. Both the slit limited spatial sampling interval (SSI) of 8.4nm and the 30m SSD are Nyquist sampled before binning and can be transferred unbinned for instrument checks or calibration.

The Thermal control System (TCS) is based on a robust passive cooling chain with margin using a single two-stage radiator with CPC reflector to cold space.

The instrument is spectro-radiometrically calibrated by using reference targets, for solar absolute radiometric calibration, and spectral calibration, and vicarious calibration via proxy targets, such as ground, moon, and atmosphere).

Figure 1.

CHIME Instrument Architecture Concept Overview

The figure below shows the conceptual layout of the instrument. It shows the main entities and building blocks of the instrument in their relative/approximate spatial and in their functional arrangement.

The platform is located at the bottom, the instrument towards the top. The instrument is mounted on the top panel of the platform, with a main mount that interfaces to the optical bench unit and auxiliary mounts that interface to the instrument support panel (ISP). They are independent, so that optical bench stiffness and ISP/Platform stiffness are decoupled.

Figure 2.

Instrument layout diagram with the two light paths for earth observation (white arrows) and calibration (sun calibration yellow, spectral calibration orange).

The instrument is closed with a cover structure equipment (black line around the main parts in the sketch) and is covered in multi-layer insulation (yellow line). The radiators of the thermal control system are mounted on the cover on the outside of the instrument. Attitude control sensors like the Star trackers, are also fully or partly on the outside.

All other components are inside the instrument cavity provided by the cover structure. The instrument features a central optical bench, the Optical Bench Unit (OBU, grey rectangle on the top of the cavity) to which all other systems interface. The front end of the optical system, the telescope with its mirrors (TELE, M1-M4) and their mounts are on the left, a part of the calibration equipment on the right. The light path (dotted line) is folded to make the instrument more compact. The Spectrometer Structure, the grey rectangle hanging on the bottom of the OBU is a structure that holds the three spectrometers, the backends of the optical system. The three identical spectrometers are in a staggered and offset arrangement. They consist of a slit, a spectrometer unit and a detector focal plane assembly (FPA). The electrical connections from the three digital read-out detectors are routed to three identical video acquisition units (VAU). Cooling for the detectors is realized by three thermal straps that interface to a common thermal link and redundant cryogenic heat pipes that connect to a focal plane radiator unit (FPA RAD) facing cold space. Cooling to the VAUs is provided by an instrument radiator unit (AUX RAD via a standard heat pipe).

In this arrangement light enters the instrument from the left through an absolute radiometric calibration mechanism that can select between sun (calibration) and earth view (normal remote sensing operation). It then is picked up by M1 of the telescope. A unit located after M4, the ICSU, switches between light coming from the telescope, spectral calibration light (provided by a spectral calibration light source), and a dark position for dark calibration. The spectrometers and their slits pick up the light from the telescope focal plane and disperse it onto the detectors where light gets converted into electrical signals.

All harness that is routed towards the Platform and the units therein (the power unit, data handling unit, control unit, and the AOCS system) has a break at a bracket at the ISP that allow easy separation of the instrument and the platform.

2.1

General Architecture

The CHIME optical system comprises a single telescope feeding three functionally identical spectrograph units. Telescope and spectrographs are design as independent units and assembled at the slit interface. The telescope images the swath onto the three slits in the telescope focal plane, each slit being integral to a spectrograph unit - i.e. the slits pick-off the subswaths from the larger focal plane of the telescope. Each spectrograph creates the spectrum of the swath in 30 m square ground samples over the full spectral range of 400 nm to 2500 nm in a single grating order. The three spectrometers are arranged such that the individual slits have a nominal 30 pixel spatial overlap. To achieve this, the slits are put in a staggered configuration with the two outer slits in a common line and the middle slit being offset. The offset translates to ~1s temporal or ~7 km spatial offset on ground

The two units, telescope and spectrographs are arranged in a folded fashion as shown in the figure below. The light enters from the left and passes the TMA. A fold mirror after telescope mirror 3 (TM3) reflects the light through the gap between TM1 and TM3. The three spectrometers are located behind TM1 and below TM3. The light from the center slit passes the slit and enters the center spectrometer directly. The left and right slit are offset, shown at the bottom of figure3, and a small fold mirror after each of the outer slits folds the optical paths over to the left and right spectrometer. In this arrangement, the focal planes meet “back to back” creating a focal plane area for a common temperature control system for the three detectors.

Figure 3.

Instrument optical path layout.

The telescope and spectrograph are designed as independent units. The three spectrograph/telescope combinations must be considered to be independent systems, though the two outer systems are nominally identical. Each spectrograph contains the stop as an aperture in close proximity to the grating. That aperture determines the stop and diffraction contribution to MTF of the spectrograph/telescope system.

2.2

Telescope Optics

Figure 4.

The telescope optics: a TMA of Wetheral-Womble type with powered mirrors TM1, TM2 and TM3, and a fold mirror at which focus correction can be applied during flight.

The telescope is a semi-classical, axis-symmetric, three mirror anastigmat (TMA) with three aparaboloidal mirrors and a fold mirror between the third powered mirror and slit plane. The clear apertures of each of the three powered mirrors (TM1, TM2 and TM3) are off axis, but the mirror surfaces are rotationally symmetric and co-axial with the telescope. It images an off-axis field into the slit plane and, because the spectrographs require slit-space telecentric feed, is designed for imagespace telecentricity over full the field of view. The telescope is designed to be faster than is strictly necessary to feed the spectrographs because that excess speed relaxes the spectrograph-to-telescope alignment tolerances, and the system stop for each of the three telescope/spectrograph pairs is located at the grating in the spectrograph.

The fold performs two functions: (a) it places the slit planes and the three spectrographs under the telescope and (b) enables the use of a refocus mechanism that can be used to compensate for change in system focal length due to moisture release from the CFRP structure in the immediate aftermath of launch. Except possibly to compensate for long-term effects, after moisture release, focus will not be required because the telescope’s focal length will be stable and the minimum orbit height is much greater than the telescope’s hyper-focal distance at all wavelengths.

Table 1.

Telescope parameters

Spot size: The telescope design produces an RMS spot radius of between 3 μm and 4 μm in the center slit, and between 3 μm and 7 μm in the outer slits. This gives a telescope that is (by design) diffraction limited at or above 895 nm over most of the field. The spot radii compare with a slit width of 60 μm.

Wavefront error: The basic requirement is to produce a sufficiently low RMS WFE across the slit plane, that the secondary field that is the image of the swath at each of the three slits, is a source giving specification compliant ISRF and ACT MTF at the spectrograph detector planes. Telescope RMS WFE current design worst case is 80.4 nm for the central slit, and 284 nm outer slits. This will be re-balanced in favor of the outer slits later.

Figure 5.

Wavefront error map: Center slit (top) with range 80.4 nm to 134 nm. Outer slits (lower) with WFE range 45 nm to 284 nm. In the outer slits, the WFE is highest at the innermost ends of the slits (light colour); in the centre slit the WFE is highest in the centre of the slit.

MTF: Along track the system’s MTF is independent of the telescope’s MTF; across track, the telescopes MTF influences the system MTF via the telescope’s sagittal geometric MTF which, to an acceptable level of approximation for preliminary assessment multiplies the spectrograph’s diffraction sagittal MTF. Apart from MTF or its corollaries, the optical performance of the telescope does not determine the system level optical performance.

The nominal design performance for the telescope in combination with the outer spectrograph is illustrated by the along-and across-track MTF in the figures below. Note that in this simulation the MTF results from the total system WFE. The MTF that results is higher than that predicted by combination of the telescope’s geometric MTF and the spectrograph’s diffraction MTF.

Figure 6.

Tangential (ALT) and sagittal (ACT) MTF at the detector’s Nyquist frequency of 13.88 cycles per millimeter, over the full detector from 400 nm to 2500 nm for the combined telescope and spectrograph for (left) the centre spectrograph unit, and (right) the outer spectrograph (where performance is lowest) for the current optical design. The asymmetry introduced by the telescope is evident in the right-hand figure.

Preliminary tolerance analysis indicates that, with the structural model refinement introduced to the mechanical model since this analysis was performed, the telescope will be sufficiently insensitive to the dominant influences on optical performance (component manufacture, alignment, gravity release and CME) that it can be assembled using existing procedures and will perform to optical specification when flown.

2.3

Spectrometer Optics

The CHIME system architecture relies on a set of three compact broadband Spectrograph Units (SU) forming the spectrometer equipment (SE). Each SU covers the complete 400-2500 nm VISIR spectral range, and have their slits arranged in a staggered configuration in the TMA focal plane. The SU consists of an all reflective optical instrument building upon AMOS’ heritage in the domain of compact, high performance, long slit spectrographs. Its customized Offner-Chrisp layout includes a key enabler feature: the de-magnification function. The original benefit behind a built-in slit de-magnification lies in the relaxation of the optical constraints put on the front-end optics, including the slit itself, through the reduction of the object space F# by an identical factor, without degrading the instrument spatial resolution. Additionally, in the specific frame of the CHIME SE design, the tailored slit de-magnification creates extra mechanical clearance between the focal plane areas of individual units through the concomitant increase of the slit length, and the associated reduction of the back-focal distance, a decisive factor for the SE staggered slit configuration.

The SU optical design is optimized to satisfy the demanding image quality and focal plane distortion requirements (smile and keystone) over the wide focal plane area through the use of two freeform mirrors. The better aberration control provided by freeform optics, over a field that is large for such compact optics, enables the design of a simple, compact, high performance SU covering the full spectral range with a single focal plane sensor. Each unit throughput is ensured by employing a squared aperture over a broadband convex diffraction grating. The SU throughput over the complete spectral range is attained by relying on an advanced grating blazing technology. The availability of a broadband diffractive component enables the design of a SU with a single Focal Plane (FP) covering the full spectral range, alleviating the use of complex optical elements and cumbersome FP co-alignment strategies. The proposed all reflective solution is fully compatible with an all-aluminum alloy build, providing a good athermal system. This solution takes advantage of an extensive manufacturing heritage at AMOS, using Single Point Diamond Turning (SPDT) technology (e.g. Sentinel 5p’s Tropomi, Proba V, Chandrayaan II).

Table 2.

CHIME SU nominal design performances

The SU design, based on the experience gained with the ELOIS [5] spectrometer, achieves the tight CHIME SE specifications in terms of both geometrical and radiometric performances. Monte Carlo analysis reveals the performances are consistent with usual tolerances.

The CHIME SU optical design is presented in the figure below. It allows the high fidelity dispersion of the 90 mm long input slit over each of the 250 contiguous spectral bands composing the focal plane. This efficient ultra-compact design builds upon AMOS’ experience taking advantage of three key elements:

Figure 7.

Top: Each CHIME SU Optical design comprises two freeform mirrors (SM1, and SM2) and a broadband, spherical, diffraction grating (GR). Bottom: The staggered arrangement of the three SUs defining the SE.

The design line spread function of the Spectrometer is shown in the figure below.

Figure 8.

Line spread function of the spectrometer along track at 2500nm.