Wide field astrophotography with a DSLR and narrow band filters

(A este artículo le tocó ser en inglés; si alguien lo necesita en español, que me diga)
Let me share my experience with this type of photography with a difficult name: Wide field astrophotography with DSLR cameras and narrow band filters. This produces unusual sights of the sky that otherwise we are used to see with the naked eye. For example: This photo is the Orion constellation. I took it from home, with a Canon 40D modified for astrophotography, a 50mm lens at f/3.5 and a total of more than 30 hours, in nine different nights. And using three narrow bands filters of 12nm bandpass. What does all this mean?



For those of us that live in a light polluted area, like in my case the very touristic “Costa Dorada” (http://www.salou-tourism.co.uk/) in Tarragona, Spain, it is almost impossible to make classical astrophotography from home. By classical I mean with a normal DSLR camera, or even with a monochrome camera and RGB filters. The reason is that the light pollution results in a significant brightness of the background sky, which swallows all but the most intense signals from the deep sky. RGB filters (including those that form the Bayer matrix in DSLR cameras)let pass a broad range of light wavelengths (for example, the red filter goes from the deep red to about the yellow) so they also let pass a lot of the light pollution that floods the light from dim sky objects. So, one possible solution is to use “narrow band” (NB) filters (there are other solutions like going to a dark place to do astrophotography, or having a remote observatory, but these are not options for me). The NB filters only let pass a very narrow (obviously as the name suggests) band of wavelengths, typically between 3 and 12 nm. Of course, very little light can go through, therefore these are very dark filters; but we could say it is a very “pure” light, almost monochrome. The manufacturers choose the wavelength that the filters let pass, usually, as a function of the types of light that are emitted by emission nebulae. These nebulae are very abundant in our Galaxy and in other galaxies, and are composed basically by hydrogen. Radiation from stars may excite hydrogen atoms, which then relax by emission of certain wavelengths (ah, quantum mechanics). The filters are therefore tuned to pick up that emission, the most common of which is by far the hydrogen-alpha line at about 653 nm, a red color. Other elements, far less abundant in space, also emit light at specific wavelengths and filters can be built to let them pass in a narrow range.

Well, so using these narrow filters it is possible to avoid the light pollution, because typical street lights emit only in certain wavelengths that do not coincide (luckily) with the typical nebulae emission lines. This is true for typical mercury and sodium lights; LED’s have a continuous spectrum and are a disaster for astrophotography, among other risks, but that’s another story. I will not explain here the details of narrow band astrophotography, and will focus on the specifics of wide field, narrow band photography for DSLR. For this purpose, DSLR cameras offer a big problem: they are color cameras; and this is done by putting color filters in front of the pixels of the sensor. Usually, there are two green pixels and one blue pixel per each red pixel. This means that if we put a hydrogen-alpha (H-alpha) filter in a DSLR camera, only one fourth of the pixels will “ see” the light from the nebulae. This results in a significant loss of signal. A DSLR camera will need much longer exposure time, at least 3 to 4 times longer, than a dedicated astronomical monochrome (no color filters) camera. For relatively “bright” objects, like M42 (Orion’s nebula) or the North America nebula, you need at least 8-10 hours of H-alpha exposure for a decent signal to noise ratio. Other filters “tuned” to other nebulae emissions, much less intense, will need even more time. That is a fact.

Focusing with most DSLR cameras is relatively easy nowadays, with the Liveview function. In wide field imaging, I assume we are using camera lenses, so we should manually pre-focus at infinity, approximately. You will see nothing through a DSLR camera visor, if you have a narrow band filter. So use the scale engraved on the lens. Then point to a medium brightness star (about mag 2 with narrow band filters, as a rough indication, actually, dimmer is better, if you can find the star in the Liveview screen), and open the Liveview. A dim star will allow you to focus more precisely, but has a problem, if you did not pre-focus well, you may not see it in the Liveview screen. If you don’t, move focus back and forth a bit: if the star is a bit out of focus it may just disappear, because the little light that passes the filter is scattered in a broad area. If you have pointed to the star well (so, if you are sure it must be “there”), zoom in as much as you can, this will help to see it. Once you have found the star, with maximum zoom, just focus as well as you can judge. You may want to recheck focus every few hours, it may drift due to temperature changes.

With wide field astrophotography, you are probably not using a big, sophisticated mount. You are probably using a simple, light mount, even one of those nice “pocket German equatorials” like Takahashi’s Teegul Skypatrol II, or the Kenko mount from Hutech. Or perhaps this interesting device, the Astrotrac. In any case, you probably don’t have a go-to system with image analysis and plate-solve capability linked to a huge star database. It means, you have to point and frame your camera by hand. Remember, with NB filters you see nothing through the visor of the camera. So one way to frame the object you want, I’d say, is to remove the narrow band filter for a moment, frame well using the visor and then put again the narrow band filter carefully. This is possible, and I have done it, but this method usually interferes with the focus. If you have focused already with the NB filter, you will probably ruin it during the filter removal and back operations. If you frame first, it may be difficult to find a good star to use it for focus in your desired field of view. Another method, that I have used most, is to focus first on a nice star, and then, frame the desired object by an iterative process: Take a short exposure of say, 20 or 30 seconds, just enough to see and identify what is in the field of view of the camera, and move by hand the pointing direction of the camera, take another short exposure, check it and so on, until you are satisfied with the framing. With DSLR-narrow band, it is normal to require several nights of exposure of the same object. So the second and third nights you have to find the same framing as the first night, by hand. Ok, with practice it is not that bad.

I’ll mention that for most nebulae you get decent signal for the H-alpha filter, and some signal for the Oxygen-II filter. The Sulphur-II filter gives very little signal, but I have used it on a few wide-field objects, like the two that I show here. By the way, in narrow band astrophotography, the colors are arbitrary. You can assign the photo from each filter to the red, to the blue or to the green channels, up to you. So that’s why the color may look not natural. They are not. See for example the photo of Orion above and this one of the Monoceros constellation, with the Rosette nebulae and the area of the Cone and Foxfur nebulae. In this case, taken with the same Canon 40D and same setup as the Orion photo, only even longer exposure, about 40 hours. Colors are totally subjective. Each color combination is called a "palette". Personal taste rules here.



I may add another entry with more info, in a few weeks. Maybe there are other aspects of wide field astrophotography with NB filters that you want to know about, just let me know to see if I can help. Let me finish saying that, although some people affirm that it makes no sense to do this type of photography because of all the difficulties, in my experience it is possible with good results. Just needs a bit extra time and patience.