AAO Newsletter July 1996 - Page 7
There have been several changes to IRIS in recent months that prospective
users should be aware of. The spectropolarimetry optics have now been corrected
for a slight focus error that delayed their commissioning. The system has
now been fully tested and appears to be working as expected. The Wollaston
prism used means we are limited to a 20" slit, but otherwise the use
of the polarimeter in this mode is very straightforward and similar to
taking normal H or K band grism spectra. The sensitivity figures are essentially
the same as those given in the IRIS manual for the grisms on their own.
We have taken the opportunity of this upgrade to revise some of the names
that positions for the wheels inside IRIS are known by. All the possible
combinations of lenses/filters/prisms/grisms etc can now be selected by
name (which hasn't been true for some time now). You should ask your support
scientist if you are unsure about any of this.
Two longstanding problems still remain. The array has a higher effective dark current now than when it was commissioned, caused by pixels being irradiated by the output amplifier, and this, rather than read noise, can be a fundamental limitation of the system performance in low background situations. We recommend when using the non-destructive read mode (method 4 in IRIS parlance) that no more than 20 NDRs be taken in a single exposure (a value between 10 and 20 is optimal for minimising both the read noise and this additional component that represents shot noise on the frame after a suitable dark - or more typically sky - frame has been subtracted). The minimum achievable effective read noise now appears to be ~ 70e- when taking NDRs. The read noise for a straight read (method 1) is still 130e-. The problem of increased noise is especially noticeable at shorter wavelengths with the IJ echelle. We would not recommend using the IJ echelle for any kind of faint object spectroscopy. In addition, there are increasing signs that the array is developing more bad pixels. There are several small patches of unresponsive pixels now scattered across the array. Again, this seems to be partially a wavelength dependent effect with more bad pixels obvious at shorter wavelengths. Secondly, the narrow band filter for the He I 10830Å line has a serious light leak and is unusable. Because of the increasingly poor response of the array at narrowband near 1µm, we do not intend replacing this filter. We are, however, purchasing a 1% wide filter centred on the 2.058µm He I line as an alternative.
A new version of the IRIS manual is currently in preparation. This will
be a Web based document, and should completely supersede the previous version.
It is hoped that this new manual will be ready within the next two months,
at which time IRIS will be 'frozen' in its current configuration with no
future developments being foreseen.
Since 2dF has already exceeded the capabilities of Autofib during a
recent commissioning run, and since Autofib is complex and therefore not
simple to maintain even when off the telescope, it has now been decommissioned.
Autofib will not be discarded however in case of an unlikely major accident
involving 2dF, or more likely, to enthrall visitors during Open Days. FOCAP
is still available for special case requests (contact the Director).
Other instruments lately decommissioned are the IPCS and the instrument Interdatas. Users of high-speed photometry will need to use the ANU software package (contact Bruce Peterson at