The design of OPTIMOS-EVE results from the scientific and technical trade-offs made during the study. Six different designs for the positioner and six different designs for the spectrograph have been studied, resulting in the adopted Phase A design.
The instrument is designed to tackle the highest priority science targets identified for the E-ELT. It is compliant (see RD5) with all the OPTIMOS-EVE Top Level Requirements (see RD3) as well as with the Requirements for OPTIMOS, a Vis-NIR Multi Object Spectrograph for the E-ELT (see AD4)[1].
The instrument consists of three main sub-systems:
1. a pick-and-place positioner
2. fibre bundles for various spectral resolutions and integral field units
3. and two highly efficient VIS-NIR spectrographs with VPH gratings working in 1st order.
A schematic overview of this system is shown in Figure 5. It is designed to fit within the volume and mass limits of the focal plane station. A more detailed description of the instrument can be found in the System Design and Analysis (RD5) and references therein. With five different observing modes the instrument is perfectly adapted to match the scientific needs.
Figure 5 Schematic overview of OPTIMOS-EVE. The 3 main subsystems of the instrument are clearly indicated: The Focal Plate Carousel - Positioner, containing 4 Focal plates with various Mono Objects Fibre inputs and IFUs as well as a robot positioner; The fibres, to transport the collected light to the spectrograph, guidance sensors and from the calibration source; The Spectrographs, consisting of a VISible and an Infrared arm, separated by a dichroic. The dispersing element is a Volume Phase Holographic Grating (VPH).
The overall complexity of OPTIMOS-EVE is comparable to that of FLAMES/GIRAFFE and the OzPoz positioner at the VLT, however there are differences: the number of different fibre bundles is 5 in stead of 3; the positioner is easier because of relaxed tolerances at the size of the E-ELT; The camera has to rotate in order to use high efficiency VPH Gratings; The instrument must be located in a cryogenic environment because the wavelength range is extended to 1.7 micron.
The fibre positioner is based on a pick-and-place design with magnetic fibre ends. It contains 4 focal plates mounted on a carousel that can be rotated. One focal plate is used for active observations on sky, while 2 robots are reconfiguring another focal plate in order to minimize reconfiguration times and allowing continuous observation with the E-ELT. More details can be found in RD5 and RD8.
Figure 6. Artist Impression of OPTIMOS-EVE on the Nasmyth platform of the E-ELT.
The Mono-Object (MO) field of view is 0.9’’ for Low and Medium Resolution and 0.81” in High Resolution. Micro lenses re-image the telescope pupil onto the fibre core, creating a very efficient light injection into the fibres. Each micro lens acts as a field lens as shown in Figure 7 (Left).
At the output side, the bare fibres are aligned to form the spectrograph entrance slit. Smaller microlenses and smaller fibre diameters are used to create a narrower slit and thereby increasing the spectral resolution (Figure 7, centre). However more fibres are needed to cover the same aperture on sky and therefore the multiplex is lower in high resolution.
The Integral Field Unit modes MI and LI have 0.3” spatial sampling on sky and operate in Low Resolution (LR) only. Table 1 summarizes the sampled strategies adopted by OPTIMOS-EVE.
Table 1. Multiplex, spectral resolution, aperture and on sky sampling for the 5 observing modes.
Observing Mode
Multiplex
Spectral Resolution
Aperture
Microlens sampling on sky
MO-LR
240
5000
0.9”
0.3’’
MO-MR
70
15000
0.9”
0.18’’
MO-HR
40
30000
0.81”
0.09’’
MI-LR
30
5000
1.8” x 3”
0.3’’
LI-LR
1
5000
7.8” x 13.5”
0.3’’
Figure 7 (Left) Fibre injection at the ELT focal plane; (Centre) slit arrangement for the MO-LR mode and MO-MR mode; (Right) The MI and LI modes.
The spectrograph concept is selected from 6 concepts in a trade-off which is a mix of meeting the top level requirements, cost, efficiency, mass, volume, modularity, technical risks and whether the components could be manufactured at all. Two identical copies of the spectrograph are required to meet the number of fibres that need to be accommodated. Each spectrograph consists of two arms: one covering the visual regime (0.37 – 0.93 μm) and one covering the near-infrared regime (0.93 – 1.7 μm).
The Spectrograph has a fast but feasible camera (F/1.86) and a very large field of view (12k spatial pixels on the detector). The design presents a high efficiency thanks to the use of VPH gratings. All optics have a reasonable size (collimator lenses fit inside a rectangle of 30x50 cm and the largest camera lens is 37 cm diameter). The camera is composed of 7 lenses, with glasses available in large and homogeneous sizes.
For the Visible Detectors we plan to use e2V CCD231 family, 6kx6k pixel format, deep depletion (baseline) or high-rho, in either case with a 'standard astro' coating or enhanced multilayer coating for improved broad-band response. The 12kx12k focal plane will be obtained with a mosaic of 2x2 chips.
For the NIR detector we plan to use HAWAII-4RG, 1.7micron cut, substrate removed devices. We use a mosaic of three chips to cover 4k (spectral) x12k (spatial). This could possibly be upgraded to 9 chips to cover 12k x 12k.
Most sub-systems of OPTIMOS-EVE will be located in the normal dome environment, like the positioner plates with carousel and robots, the positioner support frame and the fibres. Also the electronic cabinets, chiller for the cold chamber, vacuum systems and calibration box will be located in the dome environment, but thermally insulated and cooled by the water cooling system of the telescope. The spectrographs are located in a thermally controlled cold chamber filled with dry air.
Figure 8 OPTIMOS-EVE spectrograph optical design
This cold chamber is cooled by a commercial chiller that is used in industry for thermal treatment of steel constructions down to 170K, but also applied in the FMOS IR spectrograph for the Subaru telescope. This cold chamber will be cooled to 193K. Inside this cold chamber there are four cryostats located; two for the VIS channels focal plane mosaic detectors (running at 173K) and two for the NIR channels camera/detector combination (running at 120K). Those four cryostats are based on the same technique of Continuous Flow Cryostats, as more often used in VLT instruments nowadays. The four cryostats are fully independent and also the vacuum systems are separated, to prevent them from influencing each other during operation or maintenance.
Instead of using the traditional VME crate LCUs, we plan to use one or more of the following:
·National Instruments crate (either PXI platform or Compact RIO)
·Siemens Simatic S7-300 series PLC
·Beckhoff PLC - probably
For temperature control & monitoring we plan to use LakeShore units. For control of the LN2 continuous flow cryostat we plan to use the ESO TeePee. For vacuum monitoring we plan to use Pfeiffer gauges and readouts (previously known as Balzers). For enclosure control we plan to use a Polycold liquid chiller.
One of the key concepts adopted in the design of OPTIMOS-EVE control software architecture is to consider the instrument as a distributed system composed of three collaborating/communicating sub-systems: the positioner and the two spectrographs. Taking into account the requirements set in [E-TRE-ESO-586-0252, E-ELT Interfaces for Scientific Instruments] and based also on the experience gained by the team for the development of the VLT/FLAMES instrument, the baseline architecture foresees several levels of coordination software (SOS and OSes, using VLT nomenclature) and additional interacting software packages to control Instrument specific parts (ICS for the low-level part, DCS for the detectors). At the top, SOS is the only entity that dialogues with the E-ELT TCS and OH; its main responsibility is to coordinate the activities of the positioner and spectrograph specific OSes and to properly handle the archiving of the produced scientific frames. The management and coordination of the spectrograph electronics and detector sub-systems is delegated to the dedicated spectrograph OS, whereas the positioner OS has the responsibility to handle and coordinate the various activities during the plate configuration and acquisition processes.
Even though, at the very high level, the proposed design is similar compared to the VLT software model, our approach differs in the sense that we do not propose to implement the various software packages as a set of stand-alone interacting processes but to follow the component/container paradigm. The system is modelled as a set of collaborating Components located inside one or more Containers. Adopting this philosophy, we designed the overall control software architecture based on the ACS framework; the main concepts are in any case very general and may be easily adapted even in case that the E-ELT will adopt a different framework.
The resulting overall software architecture design is to some extent complex especially considering the number of software interfaces to deal with, but it easily allows to cope with some of the critical operational aspects of OPTIMOS-EVE. The necessary high degree of parallelism (spectrographs shall and will work in parallel, different field plate can be configured during an on-going observation etc.) and, on the other side, the flexibility needed to handle each sub-system independently follows naturally from the proposed architecture.
The total mass of the instrument is 20 tons and the volume is 6x6x5.7 m3. A part of the volume is allocated for electronics and for maintenance operations (see RD8).
ADC in E-ELT intermediate Focus
Many science cases of OPTIMOS-EVE would benefit of the implementation of an ADC, however for none of them is such an implementation mandatory. Several methods and positions for this Atmospheric Dispersion Corrector (ADC) have been investigated, and the best location is in the E-ELT intermediate Focus. Therefore an official “Change Request for ADC in E-ELT Intermediate Focus” E-CRE-EVE-509-1015 has been written and submitted to ESO.
The unvignetted Field of View of the E-ELT is only 5 arcminutes in diameter. The FoV can be extended to 10 arcmin, but especially laser guide star pickoff mirrors cause quite some vignetting. Access to the full field of view would be much better if these mirrors can be taken out of the field of view when GLAO is not useable. When GLAO is used, EVE will perform best if the laser asterism is fixed with respect to the field of view, rather than to the telescope.
UV transmission of the telescope is important for the efficiency down to 370nm and an extent in the wavelength range to 310nm might be also envisioned.
The principle of the technical roadmap of OPTIMOS-EVE is to keep the instrument simple and the concept low risk. Therefore we decided to adopt well-known existing technologies and components only and we do not rely on developments in the coming years, However, we are keeping an eye on technical and manufacturing advances and can incorporate them in our design at the moment when it becomes beneficial.
The Highest risk before PAE is an Interfaces change from ESO, because there is a high probability for an interface to change with a yet to be realized telescope and the cost and manpower impact of such a change can be high as well. Risk mitigation for this includes regular interaction with ESO.
The highest risks identified during the operational phase are Human Error and Earth Quakes. Other risks during operation include power dip or power failure and falling objects from a crane, but these risks apply to the whole telescope and all instruments. Specific for OPTIMOS-EVE is that due to the modularity of the system the risk for the entire instrument to fail is relatively low. Detailed risks tables can be found in RD14.
A strong link with industry is needed for a cost effective realization of both the spectrographs and the fibres. Because fibres drive the EVE system, their study is advanced and first contacts with industry have been established already during the Phase A Study. Due to the large quantities all manufacturing procedures concerning fibres and micro-lenses must be industrialized and contacts with industrial partners have been done, and no showstopper have been identified. For the spectrographs, contact with industry will be started immediately after phase A.
Further R&D after Phase A is required to demonstrate the feasibility of the HR button, especially for the assembly of small fibres with diameter smaller than 100 µm. The goal of this mode is to achieve a scrambler to obtain a good homogeneity at the fibre output. It will warrant the high accuracy required on the measures of radial velocities of objects observed (10m/s). R&D on this prototype is scheduled and financially supported for a first manufacturing during summer 2010. It will be conducted in collaboration with François Bouchy, who conducts an independent R&D on this topic.The main goal is to test it in laboratory and results are expected at the end of 2010. Further tests might be done on sky, at the OHP on the spectrograph AURELIE.
[1] The compliance matrix in RD5 shows however a slightly smaller LI than specified in AD4 without negative consequences to the Science Case.
The OPTical Multi Object Spectrograph 'Extreme Visual Explorer' for the European - Extremely Large Telescope - Consortium partners: GEPI, NOVA, RAL, NBI, INAF, AIP, LSW, IAG, LNA, ON
