Text Size:
Smaller Text Normal Text Larger Text


Follow us on Twitter

FIES: fiber-bundle constraints:

Scientific requirements

The top calibration unit
A polarisation unit
Fiber viewing with StanCam
Fiber head at the telescope
Fiber mask
Fiber run through telescope axis: rota seal and torpedo

The spectrograph
Fiber end at the spectrograph
The bottom calibration unit

Current bundle C and spare bundle B:
Design of the current bundles, and other docs

More info:
Papers describing FIES


This document is to describe the constraints imposed by the telescope and the FIES spectrograph, for a future new fiber bundle. The motivation for this document is that a new fiber bundle that delivers better radial-velocity stability is sought for. The current bundles have fibers with circular cross-sections and give a radial velocity (RV) scatter on the order of 8m/s or worse; better RV stability is expected from a bundle with octagonal fibers. To further improve the stability, a so-called double scrambler is suggested for the high-resolution fiber.

Secondly, FIES is awaiting a new polarimetry mode, for which a redesign of the fiber head and its peripherals may be necessary.
In this document, issues that relate to or are important for the new polarimetry mode are typed in red font.

Further issues that may require parts of the telescope/FIES system to be redesigned are indicated in purple font.

Scientific requirements:

  • resolutions offered: R=25000 (low-res), R=45000 (med-res), R=65000 (high-res)

  • aim for best achievable throughput for low-res and med-res modes

  • aim for best achievable RV stability for high-res

  • support for polarimetry mode: needs two adjacent med-res fibers with on-sky spacing consistent with the splitting achieved by a new polarising unit

  • support for simultaneous sky mode: needs two adjacent med-res fibers, which could be the same as the two used for the polarimetry mode

  • support for simultaneous ThAr mode: needs an additional ThAr fiber, from the fiber-exit block to the bottom calibration unit

    We have never discussed the possibility to use image slicers to improve the throughput of FIES. Image slicers allow to minimise grating losses by using a stronger lens to make the beam into the spectrograph smaller, while enhancing the spectral resolution by slicing the image of the fiber in two halves and projecting them next to each other in the 'spatial' direction. This may require to have higher cross-dispersion to allow for sufficient inter-order distance.

    Technical constraints:

    The NOT is an Alt-Az telescope with a 2.56m primary mirror and F/11.0 beam. The plate scale in the focal plane of the telescope is 7.33 arcsec/mm. The effective focal distance of the telescope is 28160 mm. See the technical details of the telescope.

    Off-axis auto-guiding is achieved in a U-shaped region around the central instrumental FOV, and is independent of the instrument in use. The available guide-field size is larger for instruments that have a smaller FOV.

    A flat tertiary mirror, at approximately 47.3 degrees tilt, can be directed into the beam of the telescope to send the light towards the FIES fiber head. The size of the elliptical pickoff mirror, major axis of 67mm and minor axis 47mm, dictates the available total FOV of 117 arcsec (16mm) diameter in which the fibers of the bundle are (to be) placed in the telescope focal plane. The center of the tertiary mirror is about 34.5--35 cm away from the telescope focal plane. See this drawing for the lay-out of the tertiary (pick-off) mirror under the telescope.

    Top calibration unit:
    At the telescope we have a calibration unit with a Halogen and a ThAr lamp. Lamp light is directed through the calibration fiber to the pivot point of a rotationally movable calibration arm. The arm, which holds two flat mirrors and a lens, projects the calibration fiber (600 micron diameter) exit onto the entrance of any of the four science fibers.

    The radius of curvature of the motion of the arm imposes that the fibers of the main bundle are placed along a curve in the focal plane of the telescope.

    In case the polarimetry mode requires a different lay-out of the fibers in the telescope focal plane, the calibration-light trajectory and mechanisms need to be redesigned, unless the fibers are placed along the current arc, and within a total distance of 16mm allowed by the pick-off mirror.

    Fiber head:
    The fiber head consist of a cylinder of 26mm diameter which holds 4 fibers. The fibers have 100 micron diameter, except for the low-resolution fiber of 200 micron diameter. The fibers are located along an arc to accommodate the rotational movement of the calibration-unit arm. The distance between the fibers is 5.7mm or 42 arcsec.

    Each fiber is located behind a small mirror (20 arcsec FOV) with a central hole. The front face of the cylinder is tilted by 87.7 degrees with respect to the axis of the cylinder, in order to reflect the light from the telescope toward StanCam for fiber viewing.

    The fiber head is placed on a V block, that itself is tilted by 4.5 degrees from vertical, to accept the folded telescope beam (pick-off mirror is at 47.3 degrees).

    In case the polarimetry mode requires a different lay-out of the fibers in the telescope focal plane, the fiber head needs to be redesigned.

    To ensure the best throughput, it was chosen to image the telescope pupil onto the fibers. The telescope is focussed on the small mirrors in front of the fibers. Bundle C has 210 micron mirror-aperture, accepting about 1.5 arcsec from the sky. The low-resolution fiber has a mirror aperture of 350 microns, accepting about 2.5 arcsec from the sky.

    Currently the order of the fibers in the head is: low-res, med-res-sky, med-res, high-res. The two med-res fibers need to be next to each other, given the current design of the fiber mask. For the new bundle the two med-res fibers need to be spaced such that the split beams of the polarimetry mode can be accommodated.

    Fiber mask:
    The mounting plate that holds the fiber-head V-block and the calibration arm, also houses the fiber mask. This sliding mask is positioned such that a particular fiber can be 'opened', while sky light is blocked from entering the other fibers. To accommodate the simultaneous-sky mode, or a future polarimetric mode, the mask can be positioned such that 2 adjacent fibers are opened to the sky.

    In case the polarimetry mode requires a different lay-out of the fibers in the telescope focal plane, the fiber mask needs to be redesigned.

    The FIES ADC allows for a FOV of about 3 arc minutes. The ADC is mounted high up in the telescope adapter, before the pickoff mirror. For a new fiber bundle there are no constraints arising from the ADC, nor vice versa.

    Polarisation unit:
    For best polarisation measurements the polarisation unit, consisting of a beam splitter and retarder plate(s), should be placed in the telescope beam before the pick-off mirror. The problem is to find a place to park the polarisation unit when not in use.

    A location high up in the adapter has been suggested, that allows the polarisation unit to use the same in-beam space as the ADC.

    The distance between the split ordinary and extraordinary beam should correspond to the distance in the telescope focal plane of two dedicated fibers in the fiber head. This may bring additional constraints on the lay-out of the fibers in the fiber head, with respect to the current situation.

    Fiber viewing:
    See this drawing for the lay-out of the tertiary (pick-off) mirror under the telescope. Under the pick-off mirror are two lenses that image the fiber onto StanCam. The lenses can only be focussed by hand mechanically, and are focussed to allow 5mm thick filters in the StanCam beam for fiber viewing.

    The fiber-viewing system poses no constraints on a new fiber bundle, other than that, if redesigned, the fiber mask holes should be large enough for fiber viewing.

    Fiber run from telescope to spectrograph:
    The fibers in the fiber bundle are protected by a flexible tube consisting of a metal coil surrounded by a soft rubber cladding. The outer diameter of the tube is 10.2 mm. The total length of the fibers and the protective tube is 42 meters.

    In the telescope's azimuth axis the bundle runs through the 'torpedo' and the 'rota seal', which allow cables and fibers and cooling water to go from the moving part of the telescope to the fixed part of the building. These two items put strong constraints on the size of the (end of the) fiber bundles. The torpedo has two dedicated cable ducts that are approximately 10x40mm in cross section. The rota seal has a dedicated cable hole of 25mm diameter, which is partly filled up with communication fibers and cables and with the two FIES bundles.

    In case both new-fiber-bundle ends are larger than the cross section of the limiting ducts, the torpedo and rota seal can be redesigned, according to our expectations of currently available required products and equipment.

    The current bundles were passed through these limiting ducts from the telescope side down towards the spectrograph building. This allowed the telescope fiber head to be its required size, but meant that the spectrograph end of the fiber-bundle had to be built to fit through these relatively small ducts.

    The beam into the spectrograph is collimated with focal length 1524mm, onto the echelle grating. The grating has a 154x306mm ruled area and is mounted at 63.5 degrees, which corresponds to an accepted beam of 154x137mm. The corresponding F-ratio of the beam with no grating losses is F/11.1.

    To allow for focal-ratio degradation in the fibers the actual beam from the fibers is narrowed by a lens that is in contact with a glass plate covering the fiber ends. The lens narrows the beam by a factor 1.75, which implies that the F-ratio of the beam exiting the fibers should be slower than F/6.4, or else grating losses due to overfilling will occur.

    The glass plate holds a 50 micron slit that covers the high-resolution fiber and the dedicated calibration fiber used for the simultaneous ThAr mode. This slit ensures the spectrograph resolution of R>60000 for these two fibers (at the expense of 40% of the light).

    The FIES camera F-ratio is F/3.0, with focal length of 520mm. With the collimator focal length of 1524mm this corresponds to a multiplication factor 2.93. Secondly, distances in the fiber exit module are apparently enlarged by the lens by a factor 1.75. As a consequence, a core-to-core spacing between fibers of 180 microns in the fiber exit block, corresponds to an echellogram spacing of 180*1.75/2.93=107.5 microns on the detector or about 8 pixels (for 13.5 micron pixels). The FWHM of 100 micron fibers is about 4.3 pixels in the cross-dispersion direction.

    Fibers of 100 micron diameter should not be positioned closer to each other than 180 microns (core to core) in the fiber exit block, in order to be able to extract simultaneous echellograms. With this distance, and with the current cross disperser, two adjacent echellograms can be recorded simultaneously bluewards from 6300A, for polarimetry, or for simultaneous sky or simultaneous ThAr modes. Redwards from 6300A the neighbouring orders of the adjacent echellograms will overlap. For larger core to core distances, the echellograms will overlap even bluer.

    The spectrograph shutter is located as close as possible to the fiber exit and lens, and is an iris shutter with 25mm diameter opening.

    Rather than a slit, one could use 70 micron fibers to obtain a spectral resolution of R>60000.

    Fiber exit block
    In the spectrograph enclosure, the fiber exit block rests on a V block, in a train together with the beam-narrowing lens and the iris shutter. Currently the cross section of this block measures 17.0 x 9.0mm in order for it to pass the rota seal and torpedo (see above) during installation of the fiber bundle.

    For the fiber exit block and a possible double scrambler on the high-res fiber, apart from the fiber-distance limitations described above under 'Spectrograph', and the spatial constraints specified under 'Fiber run from telescope to spectrograph', there are the following constraints:

    The fibers in the fiber exit block should all be covered by a single glass plate, or should be prepared such that they are all at the same focal distance to the spectrograph in one single unit.

    Currently the fiber-exit lens is located directly after (i.e. touching) the fiber exit block. The lens could be re-used for the new fiber unit. The lens diameter is 12.7mm, and has a plane front face.

    If simultaneous-ThAr mode is to be offered with the new bundle, then an additional fiber with the same size as the high-res fiber, or with the same exit slit, should be part of the fiber-exit block. This additional fiber should be spaced between the high-res and med-res fibers, such that both resolutions can be offered in simultaneous-ThAr mode.

    The two med-res fibers should be placed next to each other, to accommodate polarimetry and simultaneous sky modes.

    Currently, the order of the fibers in the fiber-exit block is: low-res, med-res sky, med-res, simulThAr-high-res, high-res. The order of the latter 4 has scientific purpose, however, the low-res fiber can be positioned on either side of the series of 4 med- and high-res fibers.

    The bottom calibration unit:
    This unit is located in the front room of the FIES building. Its purpose is to provide simultaneous ThAr light for the corresponding mode. Secondly, it provides a halogen lamp to be able to trace and flatfield the simultaneous ThAr echellogram. Thirdly, it provides access to fibers permanently aimed at the blue sky.

    In case the simultaneous-ThAr mode is to be offered with the new bundle, the simultaneous-ThAr fiber should be of sufficient length to reach this unit. But even if it is not, a different fiber could be coupled to it to obtain the required length. The current fiber couplings we use for the calibration fibers are of SMA type.

    Current bundle C and spare bundle B

    The document by Bo Lindberg describing the current optical design of the fiber-telescope and fiber-spectrograph coupling in bundles B (spare) and C.

    Papers describing FIES

    For further info on FIES, and a list of papers describing FIES, please see the commissioning document .

    John Telting, Jan 2015

  • Back to top Last modified: 21-Jan-2015