Focusing LIRIS

LIRIS is focused by moving the WHT secondary mirror up or down. The focus should be checked at the start of every night, but once a focus has been found at the beginning of the night, it is unlikely that it would need to be readjusted throughout the rest of the night. If, however, the seeing was bad at the start of the night but later improves, or if you have tight constraints concerning the PSF, then you might want to refocus the telescope. If the seeing is really bad at the start of the night, there is no point doing a focusrun and the nominal focus can be used instead, or the focus from the previous night.

Method:
The best telescope focus value for observing with LIRIS is found by taking exposures at different focus values and comparing the image seeing you get. The focus is usually within 98.10 +/- 0.05mm. Check the focus in steps of 0.05, unless the seeing is really excellent (say 0.5" or better), then a smaller step size can be used. If the seeing is not very good, you can use a larger focus step of 0.1mm instead.

Select a filter. Since no focus offset has been observed between different broad band filters, use either filter H or J to avoid high background levels in the images (however seeing improves with increasing wavelength, so H is preferable):
TO> limage h
Ask the telescope operator (OSA) to select a star of magnitude ~10-11 at low airmass and move the telescope to it
Take a sequence of images for different values of the telescope focus using the command focusrun followed by:

the number of images (i.e. number of telescope focus movements)
the exposure time, which should be at least 8-10s in order to average out seeings effects
the telescope focus value for the first image in the sequence (in mm) and
the focus increment between each image (in mm)

TO> focusrun liris 9 8 97.9 0.05

This takes nine exposures of 8 seconds, each corresponding to a different telescope focus value, starting at 97.9mm and increasing in steps of 0.05mm. These numbers will work for the majority of cases.

Select a non-saturated star (a peak count of 25 kADU is fine, but much fainter stars work fine as well). In the RTD control, switch the automatic file detection off and select Star Profile. Point the cursor at the desired star, press "b" and then "q" on the keyboard. Under Star Parameters, you will then find the current FWHM image seeing. When well focused, the stars are perfectly round, whereas defocused images show degraded PSFs, which makes the recognition of the best focus easy.
If the seeing is good (<0.8 arcsec), you will find the best focus very easily. If the seeing is bad, e.g. you see noticeable changes in FWHM in two subsequent exposures without changing focus, or no difference in two images with different focus, then don't spend too much time focusing. If you can't tell the difference between two or three different focus images, just take the value in the middle. Even if this is not the true focus, the seeing will blur the image so much that the difference is absolutely negligible.

Alternative Focus Method:
on whtdrpc1, in /home/whtguest/liris/iraf/ start an IRAF session and within the INGRID quick look reduction package (ingrid_ql), type:

in> istarfocus r559681 nim=9

and follow the instructions (i.e. mark with m one or more stars in the image, type q and wait until an xgterm window appears showing the FWHM and ellipticity plotted versus the telescope focus value (in meters, m). Finally, type q again to find the best value). NOTE: The best telescope focus value is given also in m. This method is not usually used, since the manual method is faster and straightforward.

IMPORTANT: the telescope focus is left at the last value of the sequence after the focusrun so set the telescope focus value to the best focus value you have determined:
TO> focus 98.10