Observing

During the night, observing consists of:

The following sections describe each of these steps in more detail.

Downloading a setup

If you followed the instructions in Software startup, you will already have an xterm on the rack PC to use for copying the Observing Blocks (OBs) from the GTC phase 2 web server - http://gtc-phase2.gtc.iac.es/science/Parser/sequences/ - to where the json files are stored for use by hdriver on the rack PC - /home/insuser/.hdriver/apps/. The OBs can be copied using the script get_gtc_json run in this xterm, e.g. typing get_gtc_json 01 18AGHIP 5 copies over the GTC01-18AGHIP_0005.json file.

One you have downloaded a setup, you can load it into hdriver by clicking on the Load button in the GUI. The file chooser that opens should be in the correct directory already, so you just need to select the JSON file that you just downloaded. This will load the setup, and set the correct parameters for the CCDs (e.g the binning, exposure delay etc), along with correct metadata for the FITS headers (the object name, Programme ID etc).

Taking acquisition data

There are occasions when the setup for the science data is not suitable for acquisition. For example, if the science programme consists of long exposures, or has large numbers of NSKIPS in the u-band, or if the science programme uses very small windows.

In these cases it is a good idea to change the settings to something more useful for acquisition, e.g. a full-frame setup with a few seconds exposure time. It is worth also setting the Num. Exposures to a value of 0, so acquisition images are taken continuously until you hit Stop.

See Manually setting the CCD parameters for full details of how to change the setup.

Once you are happy with the setup, select acquire from the drop-down box which sets the run type and click Start.

You can now check the pointing and focus as described in Viewing images in real-time. The command-line scripts gtc_tweak_pointing and gtc_tweak_focus are used to adjust the pointing and focus respectively. These scripts can be run from any xterm.

When adjusting the pointing, try to avoid any bad pixels or columns shown in the defect file. Important targets should also lie well away from the boundaries between CCD output channels, shown as blue lines in the real-time plot.

Taking science data

Once you are happy with the pointing and focus, Stop the acquisition run and re-load the JSON file containing the science setup. This will undo any changes you made to the CCD parameters for acquisition.

Select science from the drop-down box which sets the run type and click Start.

Quick-look data reduction

Once science data is being taken, you can start the quick-look data reduction. This allows the astronomer to monitor the cloud conditions, focus and data quality, and also measures the positions of stars in the science images for autoguiding.

Quick-look data reduction should be performed on the DRPC in a folder /home/observer/reduce/YYYY_MM_DD where YYYY_MM_DD is the date at the start of the night. This allows the reduced data to be found and archived along with the raw data at the end of the night.

The pipeline commands required for performing a quick-look reduction are described in Reducing data and plotting a light-curve. Further details of the pipeline commands and the parameters that control their behaviour can be found in the pipeline software documentation.

Guiding from science images

If you are autoguiding with the COMPO autoguider, you can ignore this section and look at Autoguiding instead.

Finally, with the quick-look reduction running, you can use a command-line script to start autoguiding from the science images. This is done from an xterm on the DRPC and it is easiest to do so from the same directory where you are performing the quick-look reduction.

From this directory, type python3 ~/scripts/autoguide.py. The script guides using the x,y positions of the stars given in the log files produced by the quick-look reduction. When prompted for the logfile, remember to give the full path to the file. If you ran this script from the /home/observer/reduce/YYYY_MM_DD this will just be the name of the log file, e.g run0012.log. If you ran the script from a different directory, you will need to enter the full path, e.g. /home/observer/reduce/YYYY_MM_DD/run0012.log.

You will also be prompted for the CCD and aperture number to use for guiding. Choose a CCD and aperture that contains a bright star, usually the reference aperture.

The last parameter you are prompted for is the averaging time. A moving average of positions from the log file is used to calculate the offsets. Enter the size of this window in seconds. This must be longer than the cadence of your data, so that the average contains at least one data point.