On-sky Fibre-Target-Alignment of the 4MOST instrument: calibration and performance

Roland Winkler, Weijia Sun, Daniel Sablowski, Thomas Liebner, Ole Streicher, Steffen Frey, Ingo Stilz, Alexander Pramskiy, Scott Smedley, Thomas Szeifert, Diogo Rio Fernandes, Gerard Zins · 2026

Read On-sky Fibre-Target-Alignment of the 4MOST instrument: calibration and performance by Roland Winkler, Weijia Sun, Daniel Sablowski, Thomas Liebner, Ole Streicher, Steffen Frey, Ingo Stilz, Alexander Pramskiy, Scott Smedley, Thomas Szeifert, Diogo Rio Fernandes, Gerard Zins (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-27882.txt.

Introduction

The 4MOST instrument will gather spectra of thousands of sky targets at once. It feeds 2436 fibres across a wide field on the VISTA telescope in Chile. Each fibre must land on its target precisely. Roland Winkler and eleven colleagues describe the fibre-target alignment calibration in 2026. The process aligns fibres with targets. Its parts include a metrology camera, the AESOP positioner, a guiding system, and projection software.

Raster scans minimise the leftover offsets between fibres and targets. One special tool was needed, a large target placed at the focal surface for calibration. About a month of calibration work was spent on the telescope. Alignment reached about twenty-four micrometres RMS error at first light. That milestone came three weeks after the final hardware was installed. Six months later, the error had fallen to about sixteen micrometres. Both figures sit far inside the requirements. Studies now aim to maximise scientific return.

Catalog record

CREATORRoland Winkler, Weijia Sun, Daniel Sablowski, Thomas Liebner, Ole Streicher, Steffen Frey, Ingo Stilz, Alexander Pramskiy, Scott Smedley, Thomas Szeifert, Diogo Rio Fernandes, Gerard Zins
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE74 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.27882

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On-sky Fibre-Target-Alignment of the 4MOST instrument: calibration and performance Roland Winkler, Weijia Sun, Daniel Sablowski, Thomas Liebner, Ole Streicher, Steffen Frey, Ingo Stilz, Alexander Pramskiy, Scott Smedley, Thomas Szeifert, Diogo Rio Fernandes, Gerard Zins Abstract The 4-metre Multi-Object Spectroscopic Telescope (4MOST) is a new wide-field, fibre-fed spectroscopic survey facility for the VISTA telescope at ESO’s Paranal Observatory. The instrument enables the simultaneous acquisition of 2436 spectra across a 4.2 deg² field of view, using a tilting spine fibre positioner feeding three dedicated spectrographs. In this paper, we describe the calibration process, and performance verification of the Fibre-Target-Alignment (FTA) process for 4MOST. We show the complete FTA process, including calibration of the individual hardware- and software components. Namely the Metrology camera system, the Fibre Positioner AESOP, a spine based Secondary Guiding System, the sky to focal surface projection software, and residual minimization via raster scans. In total, the FTA system required one special tool, a large calibration target for the focal surface, and approximately 1 month of accumulated calibration work on the telescope. The FTA process reached approx. 24 ​ μ ​ m 24\mu m ( 0.4 ′′ 0.4^{\prime\prime} ) RMS distance between fibres and targets on sky about 3 weeks after installation of the final hardware components of 4MOST, which is when 4MOST had its first light event. By the time of writing this paper, i.e. 6 months later, we reach approx. 16 ​ μ ​ m 16\mu m ( 0.27 ′′ 0.27^{\prime\prime} ) RMS. Currently, we far exceed our requirements in terms of accuracy, and are doing trade-off studies to maximize scientific returns. a]Leibniz-Institut für Astrophysik Potsdam; An der Sternwarte 16, 14482 Potsdam, Germany b]Landessternwarte Heidelberg; Königstuhl 12, 69117 Heidelberg, Germany c]Australian Astronomical Optics, Faculty of Science and Engineering, Macquarie University, NSW 2109, Australia d]European Southern Observatory, Karl-Schwarzschild-Straße, 85748 Garching, Germany keywords 4MOST, VISTA telescope, Fibre-Target-Alignment, FTA, AESOP, MCU, Secondary Guiding, Optics Characterization † † authorinfo: Further author information: (Send correspondence to Roland Winkler) R.Winkler: E-mail: [email protected] 1 THE METROLOGY SYSTEM IN VISTA/4MOST The 4MOST instrument [ 1 ] is deeply integrated in the VISTA Telescope, its components are shown in Figure 1 with the Telescope being grayed out on the right-hand side. Figure 1: Overview of the 4MOST instrument, integrated in the VISTA telescope. 3 of the 4 Metrology cameras are visible at the Spider. 1.1 FTA Hardware Components To position the fibres in the 4MOST instrument, the AESOP Fibre Positioner [ 2 ] actuates 2436 science fibres and 12 guide fibre bundles of 7 fibres each on 2448 spines. The AESOP positioner does not contain any encoders to actively control the motion of spine, instead the Metrology System is used to give a closed loop feedback. Figure 2: Left: The AESOP fibre positioner. Right: Back of the VISTA telescope with 4MOST. The Metrology System consists of 4 cameras that are mounted on the M2 spider of the VISTA telescope (see Figure 3 ), a Back-Illumination unit, and various hardware for calibration. Each camera observes the entire field of the fibre area, providing 4 independent measurements of all fibres, including 24 additional fiducial fibres which are mounted on fix metal posts that are not movable and provide a local reference frame. An image of the AESOP positioner is provided on the left-hand side of Figure 2 . Figure 3: Front view of the VISTA telescope with MetCams during installation. The fibres have 85\mu m diameter cores which are mounted in spines that have 750\mu m diameter glass ferrule tips. There is no room for separate identification markers, as for example in the MOONS instrument [ 3 ] . Instead, the 4MOST Metrology System relies on measuring the location of the fibre cores directly, which need to be illuminated from the spectrograph side. Figure 4: Left 2 panels: Ray-trace of parallel light from the sky to the focal surface of the telescope for 5 field positions. Right panel: Raytrace of the light from the focal surface to one of the metcams. The metcams view the science field through the entire optical train of the telescope, as depicted in Figure 4 . This includes the primary and secondary mirrors as well as the WFC/ADC, described in [ 4 ] , which has important consequences for calibrating the system discussed later. The opto-mechanical design of the MetCams [ 5 ] is such, that thermal expansion of various components cancel each other out such that the camera has negligible focus shift due to temperature. In addition, the cameras are water cooled to prevent any heat dissipation into the air surrounding the cameras, which would disturb science observations by introducing air turbulence. In addition, the cameras are encased in a metal shroud that prevents wind from shaking the camera body and hides shiny metal parts from the light path. The entrance opening of the MetCam is intentionally undersized, only 25mm for the 50mm optical components of the camera. This is an important feature as it sets the diffraction limit of the camera to about 2.3 pixels FWHM on the 5.5\mu m pixels of the camera detector. Figure 5: One of the lamp units is shown on the left-hand side, a schematic overview of the slit unit on the right-hand side. There are 5 Back Illumination lamps, one for each of the 3 spectrographs, secondary guiding fibres and fiducial fibres. The light is channeled by liquid light guides from the lamps to the spectrogrpahs, split into a bundle of 19 optical fibres which are then fed into the science fibres in the spectrograph slits, shown in Figure 5 . The pick-up mirror shown in the right hand side of the figure is part of the shutter mechanism and moved out of the way when observing. The back illumination light is narrow band with a wavelength of 630nm to 640nm , that is small enough to avoid dispersion effects by the telescope ADC but wide enough to prevent any speckle issues with fibres which would occur with monochromatic laser light. All metrology 4 cameras will observe the entire field of fibres [ 6 ] simultaneously, which provides 4 independent measurements for each fibre as opposed to each camera observing only a quarter of the field. The centroiding performance of 4 full field observations is the same as 4 tiled observations, as the centroiding precision is 2=\sqrt{(}4) times as good as a single camera, which is equal to the centroiding precision of a single camera observing only a quarter of the field. The implemented strategy has the advantage that seeing perturbations within the dome have less influence on the result, all cameras see all fiducial fibres, no stitching effects between tiles and the system is fault tolerant as a camera can fail without rendering the system inoperable. Figure 6: Left: Metrology Calibration Unit Right: Center of the MCU as seen from the MetCams. The footprint of each fibre on the optical component in its path to the MetCams is very small, see the top left of Figure 4 . When observing point source targets with a telescope, optical imperfections cause a slight blurring of the image, or widening of the Point Spread Function (PSF). This is somewhat tolerable for the telescope operation as atmospheric blurring (seeing)

Read the complete text