The Veloce spectrograph
Veloce
Veloce is a stabilised, high-resolution (λ/Δλ=80,000) echelle spectrograph. Veloce uses three arms to provide single-shot wavelength coverage from 396nm to 940nm:
- Rosso: 940-596nm (in use since 2018)
- Verde: 596-442nm (commissioned in July 2023)
- Azzurro:442-396nm (commissioned in July 2023)
The spectrograph is fed by a single-object fibre integral field unit, with 19 target fibres in a close-packed hexagon with a diameter of 2.5", 5 sky/background fibres offset from the target, and 2 simultaneous calibration fibres. Spatial scrambling is provided by the use of octagonal fibres. The simultaneous calibration fibres are fed by both a Menlo Systems laser frequency comb (LC) for extremely precise wavelength calibration (in the Rosso and Verde arms) and a ThXe arc lamp. Veloce is designed for bright star science, and is suitable for observing single targets brighter than I<13, V<12, B<12.
Observing
An observing guide for Veloce is available here. Veloce has a set of calibration template configurations and will accept a user input list of stars such that these stars and calibration setups can be queued up for sequential observation.
Configuration
Veloce is not configurable. It is a fixed format spectrograph. Standard calibration sequences are run at the start and end of the night, simultaneous LC or ThXe is acquired with science targets during the night, and periodic ThXe calibrations during the night. The main choices for the user is the exposure time, and whether to have one, both or no calibration fibres illuminated during the exposure.
Veloce’s standard operation mode is to use the LC during all target observations. The LC does not calibrate the Azzurro arm, and this arm is not suitable for precision radial velocity work.
An observer using Veloce for non-radial-velocity science may choose not to use the simultaneous calibration fibres as instrument stability is adequate to extract spectra for non-radial-velocity-science using periodic calibration.
Observing
All observing with Veloce is managed through the vq observing queue, which controls slewing the telescope to each new target (while reading out the previous target), acquiring guide stars, acquiring the target star onto the IFU and acquiring science exposures (and any calibrations required). Target stars can be entered into the queue using their Gaia IDs, TESS Input Catalogue (TIC) IDs, TESS TOI numbers, or entered by hand. The guide camera has a field of view of 1’, so if your targets have more than one star in a field of that size, having a 1’ finder chart will be helpful.
Exposure times and observation planning
The best photometric filters for estimating exposure times for Veloce Rosso will be R, I or TESS T magnitudes (TESS T magnitudes are essentially the same as Cousins I magnitudes for the purposes of estimating exposure times). The best bandpasses for Verde will be V or B, and the best bandpass for Azzurro will be B.
The table below provides figures of merit in AAT median seeing of 1.5” for selected wavelengths across each arm, as a function of the relevant Vega-based magnitude (for that wavelength) at magnitudes of 10, 11 and 12. Photon counts and S/N estimates are for 2 pixel bins in the spectral direction (corresponding to one spectral resolution element). The “Photons” and “S/N (phot)” column provide pure photon-counting results, while the “S/N (phot+RN)” estimates the impact of read noise when the target is spread over 19 fibres on the detector. The actual impact of read-noise will be intermediate between these two extreme cases, and will depend on seeing. (When all the light is in the 7 inner fibres, or even mostly in one fibre, a suitably weighted optimal extraction will include the effects of read-noise in many fewer pixels. When seeing is bad it will include noise from all pixels along the slit).
| Mag | λ(nm) | Exp (s) | Photons (/2pix) | S/N (phots) | S/N (phots+RN) |
| I/T=10 | 889.9 | 600 | 5465.8 | 74 | 59 |
| I/T=10 | 758 | 600 | 10196.7 | 101 | 85 |
| R=10 | 667.4 | 600 | 3079.4 | 55 | 41 |
| R=10 | 602 | 600 | 3012.7 | 55 | 41 |
| R=10 | 590.4 | 600 | 6847.9 | 83 | 67 |
| V=10 | 590.4 | 600 | 6131.4 | 78 | 63 |
| V=10 | 533.9 | 600 | 5179.2 | 72 | 57 |
| V=10 | 491.2 | 600 | 2579.4 | 51 | 37 |
| B=10 | 491.2 | 600 | 1993 | 45 | 31 |
| B=10 | 441.7 | 600 | 1248.8 | 35 | 23 |
| B=10 | 438.6 | 1800 | 1033.5 | 32 | 20 |
| B=10 | 435.4 | 1800 | 2726.5 | 52 | 38 |
| B=10 | 420.6 | 1800 | 2575.2 | 51 | 37 |
| B=10 | 412.1 | 1800 | 1369.8 | 37 | 24 |
| B=10 | 404 | 1800 | 417 | 20 | 11 |
| B=10 | 396.1 | 1800 | 60.3 | 8 | 2 |
| I/T=11 | 889.9 | 1200 | 4351.9 | 66 | 51 |
| I/T=11 | 758 | 1200 | 8118.8 | 90 | 74 |
| R=11 | 667.4 | 1200 | 2451.9 | 50 | 36 |
| R=11 | 602 | 1200 | 2398.7 | 49 | 35 |
| R=11 | 590.4 | 1200 | 5452.4 | 74 | 59 |
| V=11 | 590.4 | 1200 | 4881.9 | 70 | 55 |
| V=11 | 533.9 | 1200 | 4123.8 | 64 | 50 |
| V=11 | 491.2 | 1200 | 2053.8 | 45 | 32 |
| B=11 | 491.2 | 1200 | 1586.9 | 40 | 27 |
| B=11 | 441.7 | 1200 | 994.3 | 32 | 20 |
| B=11 | 438.6 | 2400 | 548.6 | 23 | 13 |
| B=11 | 435.4 | 2400 | 1447.3 | 38 | 25 |
| B=11 | 420.6 | 2400 | 1366.9 | 37 | 24 |
| B=11 | 412.1 | 2400 | 727.1 | 27 | 16 |
| B=11 | 404 | 2400 | 221.3 | 15 | 7 |
| B=11 | 396.1 | 2400 | 32 | 6 | 1 |
| I/T=12 | 889.9 | 2400 | 3465.1 | 59 | 45 |
| I/T=12 | 758 | 2400 | 6464.3 | 80 | 65 |
| R=12 | 667.4 | 2400 | 1952.2 | 44 | 31 |
| R=12 | 602 | 2400 | 1909.9 | 44 | 30 |
| R=12 | 590.4 | 2400 | 4341.3 | 66 | 51 |
| V=12 | 590.4 | 2400 | 3887 | 62 | 48 |
| V=12 | 533.9 | 2400 | 3283.4 | 57 | 43 |
| V=12 | 491.2 | 2400 | 1635.2 | 40 | 28 |
| B=12 | 491.2 | 2400 | 1263.5 | 36 | 23 |
| B=12 | 441.7 | 2400 | 791.7 | 28 | 17 |
(Please note that these results are from just one commissioning run to date in July 2023. They should be considered preliminary and subject to update in the future.)
You should assume an overhead or 1-2 minutes to slew to each new target, 2 minutes to acquire each new target, and an overhead of 60s for readout between subsequent exposures on each target.
Exposures of at least 600-900s are recommended for precise Doppler velocities to average over the velocity effects of short period stellar oscillations. Multiple exposures are recommended to help mitigate the effects of cosmic rays, which are especially enhanced in the deep depletion CCD used by Veloce’s Rosso arm. More information is available in the Veloce Observer's Guide.
Data reduction
A set of data reduction algorithms to extract Veloce Rosso data have been developed (and a multi-stage ADACS development process to turn those algorithm into a pipeline has commenced). As of August 2023, that system has not yet been extended to Verde and Azzurro data. This is expected by the end of 2023.
The final precision Doppler velocity system is still under development, and will be made available on Github once a first release is made, so that observers can process their data.
For more information, please contact Chris Tinney.