The throughput of UCLES has been monitored on an irregular basis since the instrument was commissioned in 1988. Typically, this has involved taking observations of a spectrophotometric standard star using a wide slit, the 79 l/mm grating, and the Tek 1K CCD, and from the measured count-rate at certain wavelengths, determining the equivalent AB magnitude of the star which would result in one photon/sec/Å at zero airmass (note that the terms "photon" and "electron" are used rather loosely here - in practice, the number of electrons liberated by a photon will depend on its energy, and the quantum efficiency of the detector. In this document, it is assumed that 1 photon = 1 electron = 1 ADU for a gain of 1 electron/ADU). A compilation of all available measurements, based on an internal AAO report (Chapman, McCowage & Ryan 1996) is given in the following table:
| Date | 3521 | 3886 | 4024 | 4419 | 5497 | 6627 | Notes* |
|---|---|---|---|---|---|---|---|
| 901114 | 17.0 | (mp) | |||||
| 1990 | 16.6 | (mp) | |||||
| 920318 | 16.4 | beam rotator in (js) | |||||
| 920324 | |
16.4 | beam rotator in (js) | ||||
| 921021 | 16.4 | (js) | |||||
| 921214 | 16.3 | (sgr) | |||||
| 9302 | primary aluminised | ||||||
| 9302 | camera folding flat realigned | ||||||
| 9304 | 16.5 | 16.6 | (js) | ||||
| 930831 | 16.4 | (js) | |||||
| 9402 | primary aluminised | ||||||
| 940615 | 16.5 | 16.3 | Feige 110 (sgr) | ||||
| 941110 | 16.2 | 16.3 | Feige 15 (sgr) | ||||
| 950214 | primary and coude 3 aluminised | ||||||
| 950415 | 16.7 | 16.7 | L970-30 (sgr) | ||||
| 9507 | echelle gratings cleaned | ||||||
| 951014 | 16.3 | 16.5 | Feige 110 (jmc) | ||||
| 9511 | coude 5 aluminised | ||||||
| 951206 | 16.7 | 16.7 | average of 3 stars (sgr) | ||||
| 9603 | coude 2 and 4 aluminised | ||||||
| 9603 | coude mirrors and beam rotator realigned | ||||||
| 960427 | 17.1 | 17.2 | 17.3 | 17.1 | data from 6 stars (jmc) 79+Tek | ||
| 000122 | 17.1 | 17.0 | 17.1 | 16.9 | HR 4468 (sdr) 79+MITLL3 | ||
| 000711 | 16.9 | 16.8 | 17.2 | 16.9 | HR 4468 (sdr) 79+Tek | ||
| 010106 | 16.3 | LTT 3864 & HR 4468 (sdr) 31+MITLL3, BR in | |||||
| 010602 | 16.3 | 16.4 | 16.6 | 16.6 | LTT 7987, HR 7596, & HR 7950 (sdr) 79+MITLL2A | ||
| 010704 | 16.7 | 16.8 | 16.9 | 16.7 | HR 4468 & HR 5501 (sdr) 79+EEV | ||
| 010710 | 16.3 | 16.7 | LTT 6248 (sgr) 79+EEV | ||||
| 010823 | 16.8 | HR 5501, HR 7596, & HR 7950 (sdr) 79+EEV @ 4510 Å | |||||
| 010824 | 16.5 | HR 5501, HR 7596, & HR 7950 (sdr) 79+EEV @ 3363 Å | |||||
| 020622 | 16.8 | 16.8 | 16.9 | HR 9087 & LTT 9239 (sdr) 79+EEV | |||
| 050418 |
16.3 |
16.6 |
16.4 |
HR 3454 (sdr) 79+EEV |
* Observers: mp=Max Pettini, js=Jason Spyromilio, sgr=Sean Ryan, jmc=Jessica Chapman, sdr=Stuart Ryder
Deriving the throughput
For reference, the procedure used to derive the throughput from the 2000 January 22 observations is now presented.
Data on HR 4468 was obtained in photometric conditions at the end of the UCLES service night, using the 79 l/mm grating, a 10" slit (the seeing was 3 arcsec!), and MITLL3 CCD. The MITLL_PLANET window was used, which provides 2048 columns (+ 40 columns of overscan) x 2496 rows. Three separate exposures were taken (one centered on the 3886/4024 Å region, one on the 5497 Å region, and another on the 6627 Å region) to minimise losses due to vignetting. The overscan was fitted and subtracted, but the data was not flatfielded. For MITLL3 data, the images have bluer orders on the left, and wavelength increases going up a column/order. The doecslit task within the echelle task of IRAF was used to trace the orders, extract signal over the full width of the profile (~60 unbinned pixels in this case), subtract off the inter-order background, and wavelength calibrate the data (note that for MITLL3 data, be sure to set "apfind.order = decreasing"). The total counts at the wavelength of interest Npeak were then measured with splot. The AB magnitude giving 1 electron/sec/Å at zero airmass then follows from:
AB = AB* + 2.5log(Ncorr)and
(Npeak * g * q)where
Ncorr = --------------- * 10^(0.4 * k * AM)
(t * d)
AB* = the AB magnitude of the spectrophotometric standard
star at the wavelength of interest (see Hamuy et al.
1992 and 1994).
g = gain of CCD (electrons/ADU), usually dependent on the
readout speed.
q = ratio of Tek quantum efficiency at the wavelength of
interest @170K, to the quantum efficiency of the CCD used at
that wavelength.
t = exposure time (seconds).
d = dispersion at the wavelength of interest (Angstroms/pixel).
k = extinction coefficient (magnitudes/airmass). For consistency
with the above results, use the canonical Hayes & Latham (1975,
ApJ, 197, 593 - Table 5) interpolated Palomar values: k(3521)=
0.609, k(3886)=0.419, k(4024)=0.368, k(5497)=0.154, k(6627)=0.093.
AM = airmass.
Stuart Ryder, sdr -@- aao.gov.au