trigger twi_flat twi 0 1jm
sched twi_flat "-t 5 -u -10 -a 90.0 -l 70.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 88.0 -l 70.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 86.0 -l 70.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 92.0 -l 70.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 94.0 -l 70.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 90.0 -l 72.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 90.0 -l 68.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 92.0 -l 68.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 94.0 -l 72.0 -e 2000.0 -i 5"
sched twi_flat "-t 5 -u -10 -a 88.0 -l 72.0 -e 2000.0 -i 5"
The first several images will probably be saturated. This is not a problem, as there are plenty to go around. We will need at least 30 to create a decent twilight flat. Each field is imaged once with a medium length (20 s) exposure, and the telescope returns to the same azimuth and elevation after at least five minutes. This allows the stars to drift so we do not image the same stars in the same pixel from image to image. I'm not sure if the azimuth should be changed from to for the southern hemisphere.
It is important that these images are taken on schedule. If they are taken late, the night sky lines will dominate and you will end up with a sky flat (see Figure 7.2) instead of a twilight flat.