Photographing Double Stars
Closeups of three photographed double stars. All three photos have the same 2’ x 2’ field of view so that relative size and brightness can be compared between the shots.
There is a lot of variation in the type of targets one can choose when doing astrophotography. The night sky itself with the Milky Way makes for an attractive target. The sun and the moon are some nearby big and bright objects that can show some incredible detail. Deeper into our solar system the planets are quite photogenic. A wealth of deep sky objects outside our solar system wait for the astrophotographer to be captured. Whether this includes large bright nebulae in our Milky Way, or far away galaxies millions of lightyears away, there is no shortage of exciting objects to go for. Double stars on the other hand are not often photographed. While they are a staple target in the visual observer’s repertoire, astrophotographers seldomly grant them much attention.
Our local astronomy club is very active with both visual and photographic amateur astronomers. One specific activity is called the ‘Object of the Month’, where we share our visual and photographic experiences with a specifically selected target that has good visibility that month. For the month of September, three double stars were selected and we specifically wanted to share both visual and photographic experiences. So I took on the challenge and tried to capture all three double stars, which turned out to be a great little project.
Astronomical League Double Star program
The selected double stars are the first three stars from the list known as the Double Star program of the Astronomical League. The list contains 100 of the most beautiful double star and multi-star systems. The requirements for observing the objects in this list are modest. A small telescope is sufficient. And conditions for double star observations don’t have to be perfect. So an ideal list for the beginning visual amateur astronomer. The program comes with a certification scheme for those that have observed all 100 objects and documented them according to specific reporting requirements.
For our little project, the first three objects of the list were selected:
Eta Cassiopeiae - RA 00h 49.1m, DEC +57° 49' - Mag. 3.5 and 7.4 - Separation 13.4"
65 Piscium - RA 00h 49.9m, DEC +27° 43' - Mag 6.3 and 6.3 - Separation 4.4"
Psi 1 Piscium - RA 01h 05.6m, DEC +21° 28' - Mag 5.3 and 5.5 - Separation 29.8"
Planning
The program from the Astronomical League is a separate database in Astroplanner, so it was easy to judge observability. The weather forecast for The Netherlands was not very promising, so I decided to run this project at my remote site in Spain. The telescope there is a Planewave CDK14 with a long focal length of approximately 2.5m. A long focal length is definitely of benefit when imaging double stars. It turned out that the targets would reach 30° altitude between 22:00 and 23:00h. So I could remotely experiment with the settings and image them late in the evening. On September 15, 2025, the conditions seemed to be right, so imaging of the three stars began.
There are a number of aspects to consider when photographing double stars.
Exposure time needs to be short. Depending on separation and relative brightness differences, optimal exposures can be short (1-5s) to ultra short (<0.1s). For a decent image, at least several tens of images need to be combined, and if using a monochrome camera, this means tens of images per channel.
The short exposure time means there are very few other stars visible. This means that automatic registration can fail and alignment has to be done manually. If so, to keep work manageable, you also don’t want to collect hundreds of frames. It also means there will not be much star field in the image. You could combine a long and short exposure, but then the brightness of the double star would be misrepresented.
You need a decent pixel-scale if you want to separate stars that are only a few arc seconds apart. All discussions about optimal pixel-size apart, the preferred pixel-scale is well under 1”/px. This can be achieved by a long focal length, or by drizzling during processing. In my case, the pixel-scale is 0.3”/px, so ideal for this kind of work.
Imaging and Processing
Photographing the double stars was done using the setup at the remote observatory. The CDK14 sits on a 10Micron GM2000HPS mount, allowing for unguided imaging. The camera used was the Moravian C3-61000Pro in combination with 2” Chroma LRGB filters. The camera is a full frame camera and the field of view is way too big for something as small as a double star. So I imaged using 1/2 the sensor. In Voyager, the imaging software, you can select subsections of the sensor as 1/2, 1/4, 1/8 etc. Using half of the sensor’s dimension in each direction makes for files that are only a quarter (30MB) of the original (120MB), which makes the data management side of things a lot easier. Using single shot exposures, the optimal exposure time was determined. Assessment of the results was done using PixInsight, as the preview function of Voyager did not seem to be well suited for that. Once an exposure was set for luminance, a sequence was created taking 60 Lum frames and 20 frames for each of the R, G and B channels. The exposure time of the R, G and B filters was set at about three times that of the Lum.
This pattern was repeated for each of the three objects. At the end a set of 25 darks was shot at 0.1s exposure time. Since a deviating geometry of the sensor was used with the ‘1/2’ setting, I wanted to make sure that the darks would be pixel-perfect with that geometry. No flats were used.
Only light processing of the files was done:
Background correction using GraXpert. Was probably not necessary, but just to be sure.
BlurXTerminator for deconvolution, also helped a bit to separate the second object, where stars were only 4” apart.
ColourCalibration. Object 1 and 3 could be registered and plate-solved so SPCC was used. Object 2 could not be plate-solved so generic ColourCalibration was applied
Automated stretch using the Statistical Stretch script from SetiAstro.
Reduce noise using NoiseXTerminator
Combining Lum and RGB channels
Results
Eta Cassiopeiae (ALDS1)
Lum: 60 * 0.3s = 18 s
RGB: 20 * 0.5s = 10s * 3 =30 s 30s
Total Exposure: 18 + 30 = 48s
A nice example of a bright blue star and a much smaller red companion star. Under normal exposure conditions, the bright blue star would completely overpower the small red star and show up essentially as a single star. With this difference in brightness, it is important that there is enough separation between the stars. In this cas it is 13”, which is sufficient.
65 Piscium (ALDS2)
Lum: 20 * 0.1 s = 2 s
RGB: 20 * 0.3 s = 6 s * 3 =18
Total Exposure: 2 + 18 = 20s
Due to the very short exposure, there are hardly any other visible stars in the image. Automatic registration failed for many frames. From the Lum frames, 20 did register. All RGB frames were registered manually using DynamicAlignment in PixInsight.
Any longer exposure would completely blend the two stars. The small separation of 4” is nicely separated here because both stars have approximately the same brightness, and exposure can be adjusted to accommodate. With a much bigger brightness difference, this would be a lot harder.
Psi-1 Piscium (ALDS3)
Lum: 60 * 1 s = 60 s
RGB: 20 * 3s = 60 s * 3 = 180 s
Total Exposure: 60 + 180 = 240 s (4min)
Two bright stars with a lot of separation (30”). This allows the image to be pushed to somewhat longer exposures. Important benefit of these longer exposures is that the surrounding star field also becomes more noticeable.
Learnings and conclusions
There is no general rule of thumb when photographing double stars. The required exposures vary greatly and can mean the difference between splitting the stars or not. But when done carefully, double stars can be quite fun objects to image. Be prepared for some manual work though, as the short exposures can prevent some of the processing tools to work properly. This type of photography is ultimately a kind of lucky imaging. When doing lucky imaging with lunar or planetary, videos are recorded rather than individual images. I considered that as well, but given the lack of information in the individual frames, I doubt that video analysing software such as Autostakkert! would know how to deal with it.
Imaging was done here using a monochrome camera with LRGB filters. When manual steps are required this means a lot of work. In such cases, photographing with an OSC camera would probably be the easier option. These photos are made using my remote setup though, which is equipped with a monochrome camera. The remote nature makes it difficult to swap out cameras for occasions like this.
The project was fun to work on, and the resulting images came out well. It made me wonder if there are double star systems that orbit fast enough so that they can be photographed over the course of a few months to a year, and the motion can be recorded. A GIF could be created showing the stars orbiting each other. If anyone knows a good candidate for this, please mention in the comments below. I would love to take on such a follow-up project.