The Dark Energy Spectroscopic Instrument (DESI) is a fiber-fed multi-object spectroscopic instrument under construction to measure the expansion history of the Universe using the Baryon Acoustic Oscillation technique.
Management of light throughput and noise in all elements of the instrument is key to achieving the high-level DESI science requirements over the planned survey area and depth within the planned survey duration. The DESI high-level science requirements flow down to instrument performance requirements on system throughput and operational efficiency. Signal-to-noise requirements directly affect minimum required exposure time per field, which dictates the pace and duration of the entire survey. The need to maximize signal (light throughput) and to minimize noise contributions and time overhead due to reconfigurations between exposures drives the instrument subsystem requirements and technical implementation.
Throughput losses, noise contributors, and interexposure reconfiguration time are budgeted, tracked, and managed as DESI Systems Engineering resources. Current best estimates of throughput losses and noise contributions from each individual element of the instrument are tracked together in a master budget to calculate overall margin on completing the survey within the allotted time. That budget is a spreadsheet accessible to the entire DESI project.
The Dark Energy Spectroscopic Instrument (DESI) is under construction to measure the expansion history of the Universe, using the Baryon Acoustic Oscillation technique and the growth of structure using redshift-space distortions (RSD). The spectra of 40 million galaxies over 14000 square degrees will be measured during the life of the experiment. A new prime focus corrector for the KPNO Mayall telescope will deliver light to 5000 fiber optic positioners. The fibers in turn feed ten broad-band spectrographs. We will describe modeling and mitigation of stray light within the front end of DESI, consisting of the Mayall telescope and the corrector assembly. This includes the creation of a stray light model, quantitative analysis of the unwanted light at the corrector focal surface, identification of the main scattering sources, and a description of mitigation strategies to remove the sources.
The Dark Energy Spectroscopic Instrument (DESI), which is currently under construction, is designed to measure the expansion history of the Universe using the Baryon Acoustic Oscillation technique. The spectra of 40 million galaxies over 14000 sq deg will be measured during the life of the experiment. A new prime focus corrector for the KPNO Mayall telescope will deliver light to 5000 fibre optic positioners. The fibres in turn feed ten broad-band spectrographs. The prime focus corrector for DESI consists of six lenses that range in diameter from 0.80 - 1.14 meters and from 83 - 237 kg in weight. The alignment of the large lenses of the optical corrector poses a significant challenge as in order to meet the fibre throughput requirements they have to be aligned to within a tolerance of ~50 micrometres. This paper details the design for the cells that will hold the lenses and the alignment and assembly procedure for the mounting of the lenses into the cells and into the complete barrel assembly. This is based on the experience obtained from the alignment of the Dark Energy Camera (DECam) instrument which was successfully assembled and aligned by the same team and we include in the paper the lessons learnt and design modifications that will be implemented on the DESI system.
The Dark Energy Spectroscopic Instrument, to be located at the prime focus of the Mayall telescope, includes a wide field corrector, a 5000 fiber positioner system, and a fiber view camera. The mapping of the sky to the focal plane, needed to position the fibers accurately, is described in detail. A major challenge is dealing with the large amount of distortion introduced by the optics (of order 10% scale change), including time-dependent non-axisymmetric distortions introduced by the atmospheric dispersion compensator. Solutions are presented to measure or mitigate these effects.
The Dark Energy Spectroscopic Instrument (DESI) is under construction to measure the expansion history of the Universe using the Baryon Acoustic Oscillation technique. The spectra of 40 million galaxies over 14000 sq. deg will be measured during the life of the experiment. A new prime focus corrector for the KPNO Mayall telescope will deliver light to 5000 fiber optic positioners. The fibers in turn feed ten broad-band spectrographs. This paper describes the mechanical integration of the DESI focal plate and the thermal system design. The DESI focal plate is comprised of ten identical petal assemblies. Each petal contains 500 robotic fiber positioners. Each petal is a complete, self-contained unit, independent from the others, with integrated power supply, controllers, fiber routing, and cooling services. The major advantages of this scheme are: (1) supports installation and removal of complete petal assemblies in-situ, without disturbing the others, (2) component production, assembly stations, and test procedures are repeated and parallelizable, (3) a complete, full-scale prototype can be built and tested at an early date, (4) each production petal can be surveyed and tested as a complete unit, prior to integration, from the fiber tip at the focal surface to the fiber slit at the spectrograph. The ten petal assemblies will be installed in a single integration ring, which is mounted to the DESI corrector. The aluminum integration ring attaches to the steel corrector barrel via a flexured steel adapter, isolating the focal plate from differential thermal expansions. The plate scale will be kept stable by conductive cooling of the petal assembly. The guider and wavefront sensors (one per petal) will be convectively cooled by forced flow of air. Heat will be removed from the system at ten liquid-cooled cold plates, one per petal, operating at ambient temperature. The entire focal plate structure is enclosed in an insulating shroud, which serves as a thermal barrier between the heat-generating focal plate components and the ambient air of the Mayall dome, to protect the seeing.
The Dark Energy Spectroscopic Instrument (DESI), currently under construction, is designed to measure the expansion history of the Universe using the Baryon Acoustic Oscillation technique. The spectra of 40 million galaxies over 14000 sq deg will be measured during the life of the experiment. A new prime focus corrector for the KPNO Mayall telescope will deliver light to 5000 fiber optic positioners. The fibers in turn feed ten broad-band spectrographs. This paper describes the overall design and construction status of the prime focus corrector. The size and complexity of the system poses significant design and production challenges. The optics of the corrector consists of six lenses, ranging from 0.8 - 1.14m in diameter, two of which can be rotated to act as an atmospheric dispersion corrector. These lenses are mounted in custom cells that themselves are mounted in a barrel assembly the alignment of which can be actively controlled by a hexapod system to micrometer precision. The whole assembly will be mounted at the prime focus of the Mayall 4m telescope at Kitt Peak observatory and will be one of the largest lens systems ever built for an optical telescope. Construction of the corrector began in 2014 and is well advanced. The system is due to be delivered to the telescope for installation in early 2018.
The Dark Energy Survey Instrument (DESI) is a 5000-fibre optical multi object spectrograph for the 4m Mayall telecope at the Kitt Peak National Observatory. Ten identical three channel spectrographs will be equipped with 500-element fibre slits. Here we focus on the architecture of the science slits and the interchangeable auxiliary slits required for calibration.
The Dark Energy Spectroscopic Instrument (DESI) is under construction to measure the expansion history of the Universe using the baryon acoustic oscillation technique. A new prime focus corrector for the KPNO Mayall telescope will deliver light to 5,000 fiber optic positioners feeding ten broad-band spectrographs. The positioners have eccentric axis kinematics. Actuation is provided by two 4mm diameter DC brushless gear-motors. An attached electronics board accepts a DC voltage for power and CAN messages for communications and drives the two motors. The positioner accepts the ferrulized and polished fiber and provides a mechanically safe path through its internal mechanism. Positioning is rapid and accurate with typical RMS errors of less than 5 μm.
The Dark Energy Spectroscopic Instrument (DESI) is under construction to measure the expansion history of the Universe using the Baryon Acoustic Oscillation technique. The spectra of 40 million galaxies over 14,000 sq. deg. will be measured during the life of the experiment. A new prime focus corrector for the KPNO Mayall telescope will deliver light to 5000 fiber optic positioners. The fibers in turn feed ten broad-band spectrographs. We describe the ProtoDESI experiment, planned for installation and commissioning at the Mayall telescope in the fall of 2016, which will test the fiber positioning system for DESI. The ProtoDESI focal plate, consisting of 10 fiber positioners, illuminated fiducials, and a guide, focus and alignment (GFA) sensor module, will be installed behind the existing Mosaic prime focus corrector. A Fiber View Camera (FVC) will be mounted to the lower surface of the primary mirror cell and a subset of the Instrument Control System (ICS) will control the ProtoDESI subsystems, communicate with the Telescope Control System (TCS), and collect instrument monitoring data. Short optical fibers from the positioners will be routed to the back of the focal plane where they will be imaged by the Fiber Photometry Camera (FPC) or back-illuminated by a LED system. Target objects will be identified relative to guide stars, and using the GFA in a control loop with the ICS/TCS system, the guide stars will remain stable on pre-identified GFA pixels. The fiber positioners will then be commanded to the target locations and placed on the targets iteratively, using the FVC to centroid on back-illuminated fibers and fiducials to make corrective delta motions. When the positioners are aligned with the targets on-sky, the FPC will measure the intensities from the positioners’ fibers which can then be dithered to look for intensity changes, indicating how well the fibers were initially positioned on target centers. The final goal is to operate ProtoDESI on the Mayall telescope for a 6-hour period during one night, successfully placing targets on the intended fibers for the duration of a typical DESI science exposure.
The Dark Energy Spectroscopic Instrument (DESI) is a Stage IV ground-based dark energy experiment and will be used to conduct a five year survey covering 14,000 deg2 to z=3.5. This survey is accomplished using five thousand robotically positioned optical fibers that can be quickly reconfigured with a 5 μm positioning accuracy. The fiber performance in the near and far field of two types of robotic positioners are currently being investigated: tilting spine mechanical simulators and eccentric axis (or θ-φ) positioners. The far field performance of the fiber is important since the instrument efficiency is adversely affected if light from the fibers enters the spectrograph at a faster focal ratio than the spectrograph can accept (f/3.57 in the DESI design). This degradation of the focal ratio of light is caused by light entering the fiber off axis (tiliting positioner) or bending, twisting, and stress of the fiber (eccentric axis) positioner. The stability of the near field intensity distribution of the fiber is important since this determines the spectrograph point spread function (PSF). If the PSF changes from the calibration to the science exposures, this will result in an extraction bias. For DESI, a particular concern is the distortions in the PSF due to movement of the fibers during
We describe the fiber system of the Dark Energy Spectroscopic Instrument (DESI). Its primary science goal is to provide
a survey of 14,000 square degrees of the extragalactic sky using the Mayall 4m telescope in five years. The fibre system
will provide a multiplex gain of 5000 so that more than 20 million galaxies can surveyed. Applying a number of tests to
the survey dataset should allow the evolution of the equation of state of the universe to be determined to greater accuracy
than before. The fibre system will provide a multiplex gain of 5000 with very high levels of performance.
The KPNO Nicholas U. Mayall 4-meter telescope is to be the host facility for the Dark Energy Spectroscopic Instrument (DESI). DESI will record broadband spectra simultaneously for 5000 objects distributed over a 3-degree diameter field of view; it will record the spectra of approximately 20 million galaxies and quasi-stellar objects during a five-year survey. This survey will improve the combined precision of measurement on the dark energy equation of state today (w0) and its evolution with redshift (wa) by approximately a factor of ten over existing spectroscopy baryon acoustic oscillation surveys (e.g., BOSS1) in both co-moving volume surveyed and number of galaxies mapped. Installation of DESI on the telescope is a complex procedure, involving a complete replacement of the telescope top end, routing of massive fiber cables, and installation of banks of spectrographs in an environmentally-controlled lab area within the dome. Furthermore, assembly of the instrument and major subsystems must be carried out on-site given their size and complexity. A detailed installation plan is being developed early in the project in order to ensure that DESI and its subsystems are designed so they can be safely and efficiently installed, and to ensure that all telescope and facility modifications required to enable installation are identified and completed in time.
BigBOSS is a proposed ground-based dark energy experiment to study baryon acoustic oscillations (BAO) and the
growth of structure with a 14,000 square degree galaxy and quasi-stellar object redshift survey. It consists of a 5,000-
fiber-positioner focal plane feeding the spectrographs. The optical fibers are separated into ten 500 fiber slit heads at the
entrance of ten identical spectrographs in a thermally insulated room. Each of the ten spectrographs has a spectral
resolution (λ/Δλ) between 1500 and 4000 over a wavelength range from 360 - 980 nm. Each spectrograph uses two
dichroic beam splitters to separate the spectrograph into three arms. It uses volume phase holographic (VPH) gratings for
high efficiency and compactness. Each arm uses a 4096x4096 15 μm pixel charge coupled device (CCD) for the
detector. We describe the requirements and current design of the BigBOSS spectrograph. Design trades (e.g. refractive
versus reflective) and manufacturability are also discussed.
BigBOSS is a Stage IV dark energy experiment based on proven techniques to study baryon acoustic oscillations and the growth of large scale structure. The 2010 Astronomy and Astrophysics Decadal Survey labeled dark energy as a key area of exploration. BigBOSS is designed to perform a 14,000 square degree survey of 20 million galaxies and quasi-stellar objects. The project involves installation of a new instrument on the Mayall 4m telescope, operated by the National Optical
Astronomy Observatory. The instrument includes a new optical widefield corrector, a 5,000 fiber actuator system, and a multi-object spectrometer. Systems engineering flowdown from data set requirements to instrument requirements are discussed, along with the trade considerations and a pre-conceptual baseline design of the widefield optical corrector, spectrometer and fiber positioner systems.
BigBOSS is a proposed ground-based dark energy experiment to study baryon acoustic oscillations (BAO) and the
growth of large scale structure. It consists of a fiber-fed multi-object spectrograph designed to be installed on the Mayall
4-meter telescope at Kitt Peak, Arizona. BigBOSS includes an optical corrector assembly and 5000-fiber-positioner
focal plane assembly that replace the existing Mayall prime focus hardware. 40-meter long optical fiber bundles are
routed from the focal plane, through the telescope declination and right ascension pivots, to spectrographs in the
thermally insulated FTS Laboratory, immediately adjacent to the telescope. Each of the ten spectrographs includes three
separate spectral bands. The FTS Laboratory also houses support electronics, cooling, and vacuum equipment. The
prime focus assembly includes mounts for the existing Mayall f/8 secondary mirror to allow observations with
Cassegrain instruments. We describe the major elements of the BigBOSS instrument, plans for integrating with the
Telescope, and proposed modifications and additions to existing Mayall facilities.
We describe the fiber optics systems for use in BigBOSS, a proposed massively parallel multi-object spectrograph
for the Kitt Peak Mayall 4-m telescope that will measure baryon acoustic oscillations to explore dark energy.
BigBOSS will include 5,000 optical fibers each precisely actuator-positioned to collect an astronomical target’s flux
at the telescope prime-focus. The fibers are to be routed 40m through the telescope facility to feed ten visible-band
imaging spectrographs. We report on our fiber component development and performance measurement program.
Results include the numerical modeling of focal ratio degradation (FRD), observations of actual fibers’ collimated
and converging beam FRD, and observations of FRD from different types of fiber terminations, mechanical
connectors, and fusion-splice connections.
The BigBOSS instrument is a proposed multi-object spectrograph for the Mayall 4m telescope at Kitt Peak, which will
measure the redshift of 20 million galaxies and map the expansion history of the universe over the past 8 billion years,
surveying 10-20 times the volume of existing studies. For each 20 minute observation, 5000 optical fibers are
individually positioned by a close-packed array of 5000 robotic positioner mechanisms. Key mechanical constraints on
the positioners are: ø12mm hardware envelope, ø14mm overlapping patrol zones, open-loop targeting accuracy ≤ 40μm,
and step resolution ≤ 5μm, among other requirements on envelope, power, stability, and speed. This paper describes the
design and performance of a newly-developed fiber positioner with R-θ polar kinematics, in which a flexure-based linear
R-axis is stacked on a rotational θ-axis. Benefits over the usual eccentric parallel axis θ-φ kinematic approach include
faster repositioning, simplified anti-collision schemes, and inherent anti-backlash preload. Performance results are given
for complete positioner assemblies as well as sub-component hardware characterization.
The BigBOSS experiment is a proposed DOE-NSF Stage IV dark energy survey. The all sky survey will be used
to study the baryon acoustic oscillation (BAO) and growth of large scale structure from 0.2 < z < 3.5. Key to
the timely success of BigBOSS is the total optical throughput of the system. The guide, focus/alignment system
will provide essential pointing information,
eld acquisition, atmospheric monitoring and alignment corrections
all used to maximize light throughput.
BigBOSS is a proposed DOE-NSF Stage IV ground-based dark energy experiment designed to study
baryon acoustic oscillations (BAO) and the growth of large scale structure with a 14,000 square
degree survey of the redshifts of galaxies and quasi-stellar objects. The project involves
modification of existing facilities operated by the National Optical Astronomy Observatory
(NOAO). Design and systems engineering of a preliminary 3 degree field of view refractive
corrector, atmospheric dispersion corrector (ADC), and 5000 actuator fiber positioning system are
We describe a non-contact optical measurement method used to determine the surface flatness of a cryogenic sensor
array developed for the JDEM mission. Large focal planes envisioned for future visible to near infra-red astronomical
large area point-source surveys such as JDEM, WFIRST, or EUCLID must operate at cryogenic temperatures while
maintaining focal plane flatness within a few 10's of μm over half-meter scales. These constraints are imposed by
sensitivity conditions that demand low noise observations from the sensors and the large-field, fast optical telescopes
necessary to obtain the science yield. Verifying cryogenic focal plane flatness is challenging because μm level
excursions need to be measured within and across many multi-cm sized sensors using no physical contact and while
situated within a high-vacuum chamber. We have used an optical metrology Shack-Hartmann scheme to measure the
36x18 cm focal plane developed for the JDEM mission at the Lawrence Berkeley National Laboratory. The focal plane
holds a 4x8 array of CCDs and HgCdTe detectors. The flatness measurement scheme uses a telescope-fed micro-lens
array that samples the focal plane to determine slope changes of individual sensor zones.
Mission concepts for NASA's Wide Field Infrared Survey Telescope (WFIRST)1,2, ESA's Euclid3,4 mission, as well as
next-generation ground-based surveys require large mosaic focal planes sensitive in both visible and near infrared (NIR)
wavelengths. We have developed space-qualified detectors, readout electronics and focal plane design techniques that
can be used to intermingle CCDs and NIR detectors on a single, silicon carbide (SiC) cold plate. This enables optimized,
wideband observing strategies. The CCDs, developed at Lawrence Berkeley National Laboratory, are fully-depleted, pchannel
devices that are backside illuminated and capable of operating at temperatures down to 120K. The NIR
detectors are 1.7 μm and 2.0 μm wavelength cutoff H2RG® HgCdTe, manufactured by Teledyne Imaging Sensors under
contract to LBNL. Both the CCDs and NIR detectors are packaged on 4-side abuttable SiC pedestals with a common
mounting footprint supporting a 44 mm mosaic pitch. Both types of detectors have direct-attached readout electronics
that convert the detector signal directly to serial, digital data streams and allow a flexible, low cost data acquisition
strategy to enable large data rates. A mosaic of these detectors can be operated at a common temperature that achieves
the required dark current and read noise performance necessary for dark energy observations. We report here the
qualification testing and performance verification for a focal plane that accommodates a 4x8 array of CCDs and HgCdTe
Fully depleted, back-illuminated, p-channel CCDs developed at Lawrence Berkeley National Laboratory exhibit high
quantum efficiency in the near-infrared (700-1050nm), low fringing effects, low lateral charge diffusion (and hence
small, well-controlled point spread function), and high radiation tolerance. Building on previous efforts, we have
developed techniques and hardware that have produced space-qualified 4-side abuttable, high-precision detector
packages for 10.5μm pixel, 3.5k x 3.5k p-channel LBNL CCDs. These packages are built around a silicon carbide
mounting pedestal, providing excellent rigidity, thermal stability, and heat transfer. Precision fixturing produces
packages with detector surface flatness better than 10μm P-V. These packages with active areas of 36.8mm square may
be packed on a detector pitch as small as 44mm. LBNL-developed Front End Electronics (FEE) packages can mount
directly to the detector packages within the same footprint and detector pitch. This combination, along with identically
interfaced NIR detector/FEE packages offers excellent opportunities for high density, high pixel count focal planes for
space-based, ground-based, and airborne astronomy.
Optical testing of large mirrors for space telescopes can be challenging and complex. Demanding optical requirements
necessitate both precise mirror figure and accurate prediction of zero gravity shape. Mass and packaging constraints
require mirrors to be lightweighted and optically fast. Reliability and low mass imply simple mounting schemes, with
basic kinematic mounts preferable to active figure control or whiffle trees. Ground testing should introduce as little
uncertainty as possible, ideally employing flight mounts without offloaders. Testing mirrors with their optical axes
horizontal can result in less distortion than in the vertical orientation, though distortion will increase with mirror speed.
Finite element modeling and optimization tools help specify selective reinforcement of the mirror structure to minimize
wavefront errors in a one gravity test, while staying within mass budgets and meeting other requirements. While low
distortions are necessary, an important additional criterion is that designs are tolerant to imperfect positioning of the
mounts relative to the neutral surface of the mirror substrate. In this paper, we explore selective reinforcement of a 2-meter class, f/1.25 primary mirror for the proposed SNAP space telescope. We specify designs optimized for various
mount radial locations both with and without backup mount locations. Reinforced designs are predicted to have surface
distortions in the horizontal beam test low enough to perform optical testing on the ground, on flight mounts, and
without offloaders. Importantly, the required accuracy of mount locations is on the order of millimeters rather than
tenths of millimeters.
SNAP is a proposed space-based experiment designed to study dark energy and alternate explanations of the acceleration
of the universe's expansion by performing a series of complementary systematic-controlled astrophysical measurements.
The principal mission activities are the construction of an accurate Type Ia supernova Hubble diagram (the supernova
program) and conducting a wide-area weak gravitational lensing (WL) survey. WL measurements require highly
constant point spread function (PSF) second moments (ellipticity), and the aim of this study is to expand on the 2005
Sholl, et al. preliminary work, specifically via use of the Ball Aerospace integrated modeling tool, EOSyM (End-to-end
Optical System Model). This modeling environment combines thermal, structural and optical effects, including
alignment errors, manufacturing residuals and diffraction, in an integrated model of the telescope. Thermo-mechanically
induced motions and deformations of the mirrors are modeled as well as other disturbances, and corresponding ellipticity
variations of the PSF are quantified for typical operational scenarios. In this study, the effects of seasonal variations in
solar flux, transients introduced when pointing the body-fixed Ka-band antenna toward Earth, 90° roll maneuvers
(planned every three months of operations) and structure dimensional changes associated with composites desorption are
quantified and introduced into the optical system. Uncertainty in the telescope ellipticity distribution may be reduced by
examination of foreground stars within the field of view. Reference is made to ongoing work on the use of foreground
stars in quantifying the PSF.
SNAP is a proposed space-based experiment designed to study dark energy and alternate explanations of the acceleration of the universe's expansion by performing a series of complementary systematic-controlled astrophysical measurements. The principal mission activities are the construction of an accurate Type Ia supernova Hubble diagram (the supernova program), and conducting a wide-area weak gravitational lensing (WL) survey. WL measurements benefit from a highly constant point spread function (PSF). The goal of this study is to quantify the anticipated variations in PSF arising from on-orbit thermal variations and and shrinkage associated with dryout of the composite telescope metering structure. A
combined thermo-mechanical-optical analysis tool was developed, and WL metrics whisker and effective anisotropy quantified for thermal and composite structure dryout effects. Stability limits necessary for WL are defined, and compared to stability tolerances defined for the supernova program. The mission is designed for operations at at the Earth-Sun L2 Lagrange point, where thermal disturbances from Earth are minimal. In this study, the effects of seasonal variations in solar flux, transients introduced when pointing the body-fixed Ka-band antenna toward Earth and 90° roll maneuvers (planned every three months of operations) are quantified, and introduced into the optical system. Whisker and effective anisotropy were computed, and found to be well below the WL requirement for stability. The effects of
composite structure shrinkage due to on-orbit H2O desorption are discussed, and estimated to be below WL limits for
daily observations, at the beginning of the WL phase of the mission.
We present the baseline telescope design for the telescope for the SuperNova/Acceleration Probe (SNAP) space mission. SNAP’s purpose is to determine expansion history of the Universe by measuring the redshifts, magnitudes, and spectral classifications of thousands of supernovae with unprecedented accuracy. Discovering and measuring these supernovae demand both a wide optical field and a high sensitivity throughout the visible and near IR wavebands. We have adopted the annular-field three-mirror anastigmat (TMA) telescope configuration, whose classical aberrations (including chromatic) are zero. We show a preliminary optmechanical design that includes important features for stray light control and on-orbit adjustment and alignment of the optics. We briefly discuss stray light and tolerance issues, and present a preliminary wavefront error budget for the SNAP Telescope. We conclude by describing some of the design tasks being carried out during the current SNAP research and development phase.