Type afwake to wake AUTOFIB up. Technical staff should already have done a load_module at the beginning of the run (this checks and updates the positions of all usable fibres [ie. those fibres not dead or disabled]), but the observer or SA should check to ensure it has been done. When you wake AUTOFIB up, you should hear it `doing its exercises' as it opens and closes the gripper jaws a few times. You are now ready to set-up fields.
Test as many of the program fiber setups as you can during the afternoon, with af2_setup. To set-up a field, transfer the .cfg file from the SPARC to disk$lpvf_vms:[obstest.af2]:
This directory should also contain the latest wht_prime.dat (distortion map information) and large_fibres.dat (current fibre configuration) files. (N.B. You should already have confirmed that the .cfg file is a legal configuration based on the latest fibre information in the wht_prime.dat and large_fibres.dat files, by running it through the af2_configure program on the mountain-top SPARCs.
When you are ready to start the set-up, type afsetup fieldname (where the field name is fieldname.cfg. It will then ask to ensure that the TV gain is down and then prompt you for the siderial time at which the field is to be observed. If you are setting up in the afternoon - put in the LST closest to the hour angle that you configured the field for. It will take a few seconds to process the field, and then it will start to move fibres. It takes about 10 secs/fibre and so a full configuration of 120 fibres will take approx 20 minutes.
Time-Saving Tip!!! If you just want to *test* a setup (ie. in the afternoon), and don't want to actually move all of the fibres, move the fixed (off-axis) probe into the field (by typing afmprob 150000) before typing afsetup fieldname. The AF2 micro will still process the field, and you will know if your field is okay (ie. if it can be setup on the sky). However, the micro will not be able to actually move the fibres, since it will recognize that the fixed probe is in the way. You will get an error message to that effect (which you can ignore), and you can then go on to test the next field. This will save you a lot of time ... remember to move the probe out of the field (afmprob 0) before you want to do an actual setup on the sky.
If you want, you can also put the fibres in a ring configuration during the afternoon (the af2_allcircle command) to get light from an arc lamp, or a dome flat-field (the most even illumination of the fibres comes from a torch shone onto the Naysmyth flat in the GHRIL room). A torch flat will give you some idea of the relative transmission of the fibres - just in case there are any that have drastically decreased in throughput since the last run. You can run the transm routine in the WYFFOS data reduction package to give you the relative throughputs. An up-to-date list of the fibre throughput should be kept in the AUTOFIB folder in the control room, and is also available **here**. Make sure that any low fibre transmissions are NOT caused by a poor connector. This will be obvious from the fact that all the fibres in a bundle will be low. (Each bundle contains 9 fibres and the relationship between bundle and pivot number should be in the WYFFOS folder in the control room, and also **here**). If a bundle is low, ask the duty technician to re-make the bundle. When you are finished doing setups, or if AF2 will not be used for a while, AUTOFIB should be sent to sleep: afsleep.