Really good astrophotography requires long integration times. Many hours of data go into a good narrowband image. In the UK we do not get many clear nights, and when it is clear, the astro-imager still needs to get some sleep. As the astronomer is the telescope’s least reliable component, a lot of these problems are solved by removing him from the equation as much as possible....

Most people dream of a fully robotic telescope imaging system. A flawless integrated system that detects when the sky is clear, opens the dome, points the telescope and frames the shot. It seamlessly does all the imaging and calibration frames, and then switches itself off in the morning. Whilst this would be a delightful place to be, it isn’t terribly practical for the average budget. You need a lot of decent equipment and clever software to manage the whole process. If you spend £8k on a Paramount ME, it comes with a fantastic suite of software that does all these things, presuming you have a motorised dome, and a motorised focuser and a motorised everything else. And a clear sky detector. And a rain alarm.

As you have figured by now there are a large number of issues that you have to solve before you can do true automatic robotic DSO imaging. For most of use, the financial and time commitment to get it all working is beyond us.

I believe that more astrophotographers need to embrace a simplified variant of this – all night unattended imaging. If you can take an hour of autoguided 5 minute sub exposures, then you probably have enough equipment to do all-night unattended imaging.

All-night unattended imaging simply means setting up the telescope and camera on the target and collecting data as you normally do. Instead of switching everything off and going to bed when you’ve had enough, you go to bed early, and leave it gathering data on that target for as long as possible. You get massive integrations, which make for good images, and you also get some sleep!


There are a couple of issues that you need to have nailed to do this:

  • Autoguiding. You need a reliable autoguiding system that will keep working for hours at a time. If it works for one hour, chances are it will work for 6.
  • Power – you need to be running of mains power, not batteries.
  • Automatic taking pictures. All astronomical CCD cameras take their exposures from computer control. Some DSLRs need a bit of extra work. You just need to computer to keep snapping frames all night.
  • Weather. You need it not to rain. I’m not proposing a motorised shed roof linked to a rain alarm (although you can if you want to), just a sensible look at the weather forecast beforehand. In my part of the UK it is usually easy to tell at bedtime if it is likely to rain in the next 6 hours. A light shower don’t harm much anyway.
  • Correct target. You need to start imaging something that the telescope can track for a long time before crashing into the pier. Something on the up in the east.
  • Switch off. You need the whole system to switch off before the telescope crashes into the pier. The easiest way of doing this is to get a timer plug from a DIY shop and set it to turn off the power at a preset time – e.g. 4am.

      How do you know what time the telescope will crash into the pier? It isn’t a simple calculation, so Work it out empirically as follows
    • GOTO the scope at the target, and note the RA the scope controller says it is pointing at. E.g. 16h 21min.
    • Using the scope controller (e.g. Autostar) slew the scope west, past the meridian and keep going until just before a collision occurs.
    • Look a the RA on the controller. It might now say 22h 45min. Take the difference. The scope can therefore track the sky for about 6 hours and 20mins. Round this down a bit.
    • What is the time now? Say it is 10pm. Set the timer to turn off the scope at 4am.
    • Bear in mind you might want to set it earlier if it is going to get light.
    • GOTO back to the target and frame the shot.


That is pretty much it. Do not stay up late on a week night thinking “just one more sub-exposure” and get reduced sleep – go to bed at normal time and let the scope work all night. You need to carefully choose your target. Perhaps image one target in the evening before bed time, and then set the telescope on a new easterly target that it can work on all night. For example, I can set the telescope on the Heart Nebula region at 10pm and leave it taking 900s sub exposures until 5am. In the morning I have 6 hours of data. Do this on three nights with three narrowband filters, and you’ve got the data for a great image.

This sort of approach doesn’t work out of the box – you need to iron the wrinkles out of the system. Bring your own variations to the mix depending on your equipment and skills. I’m a programmer during the day job, so I’ve written a program that allows me to set the shutdown time on the observatory computer. When this time is reached, the program closes the imaging and guiding software, and issues a park command to the scope. All I have to do in the morning is kill the power to the observatory and pull the roof shut.

It always amazes me that astrophotographers are happy to go through all the hassle of setting up for imaging, and then turn it off after 2 hours because they have to go to bed - they haven't really thought it though, have they?

Too many clear nights are wasted by sleepy astronomers – start doing all-night imaging and the quality of your images will increase dramatically.