AAQ Astrophotography Section
An Introduction To Astrophotography
First StepsAstrophotography is a demanding but rewarding pursuit – demanding because to achieve excellence requires a number of factors to all be at the highest level, and rewarding for the satisfaction of having achieved success in a difficult field. It is also rewarding because a permanent record is created which can be shown to others, often of something not directly visible except maybe as a grey smudge. The entry level into this pursuit need not be high if one is happy to photograph the Moon and the brightest planets. Such objects should not be decried – there are some renowned amateur astrophotographers who photograph little else, even travelling to another continent to seek excellent “seeing” for high resolution photographs. Although not ideal, a Dobsonian telescope can be used for lunar photography, as it is very stable and virtually vibration-free. Apart from a telescope, mount and SLR camera, all that is required is a camera adaptor and some film. ISO 200 or 400 is ideal as it will result in short exposures and it is reasonably fine-grained.
An aspect needing resolution is the actual focal ratio of an eyepiece projection setup. There is a formula that can be applied, but this needs knowledge of dimensions of not-easily-established elements of the optical system. Another way is to use photography as follows. First photograph the moon at prime focus. (It is assumed that the focal ratio of the telescope is known or can be readily established.) Next, photograph an area of the moon with eyepiece projection using each eyepiece in turn. Develop all images, mount them in slide mounts and project them on to a piece of paper taped to a wall or window. By measuring the distance between the same craters on the projected images, the focal ratios of the eyepiece projection setups can be calculated, as image size is proportional to focal ratio. Whether or not one knows the focal ratio for certain, it is still advisable to bracket exposures, as most subjects are not uniformly illuminated. Another avenue to pursue – one not needing a telescope – is long exposure wide field photography using a tracking platform. The construction of such a device has been well described in numerous publications (eg Astronomy, September 1984) and on Internet sites and will not be attempted here. With careful use, good wide-angle photos of constellations using normal 35 to 105 mm lenses can be achieved with exposures of 5 to 10 minutes, depending on the film used. Films And FiltersFilms Films for astrophotography must satisfy a number of criteria. They must be fast, to record faint objects within a reasonable time; fine grained to avoid a grainy background and to achieve small star images; and they must be sensitive to red light at 656 nanometres wave length to record emission nebulae. The rated speed of a film is not necessarily a good indicator of its speed relative to other films for long exposures as the following examples, based on research by Robert Reeves, show.
For colour slides, Kodak Elite Chrome 200 is a good choice. This film has good red sensitivity, which leads to spectacular nebula photos. It should not be hypered but it can be push processed. A one-stop push gives an effective speed of about 320 ISO. An exposure of between 40 and 60 minutes at f6 will result in excellent deep-sky images. If your choice is colour prints, the best readily available films are Kodak Royal Gold 400 and Fuji Superia 800. Both have adequate sensitivity to the H-alpha wavelength and yield excellent results with an exposure of 40 to 60 minutes at f6. Now that Kodak has discontinued production of Technical Pan film, there is no black-and-white film suitable for deep-sky astrophotography, although most films would be acceptable for photographing solar system objects. Filters Most filters find their place in visual astronomy but some are useful for astrophotography. A broad band light rejection filter such as Lumicon’s Deep Sky Filter is very useful. This will reduce sky glow from natural and artificial sources and can be used with both black and white and colour films. It allows a film exposure from a suburban site to be extended from about 10 minutes to about 45 minutes before the image starts to get washed out by the bright sky. However, it is now relatively expensive. If using a black-and-white film from a suburban site to image deep-sky objects such as galaxies or globular clusters, a red filter such as a Wratten number 25 will allow long exposures, but at the expense of loss of a significant portion of the light needed for the image. Without filters, astrophotography of deep-sky objects can successfully be undertaken only at reasonably dark sky sights. Appropriate filters will allow quite good results to be obtained from an average backyard in the suburbs, thereby allowing the astrophotographer to hone skills at home before that important, infrequent excursion to a dark sky location. A side issue here relates to consumer digital cameras. At least some of these are known to have a filter in front of the ccd or cmos detector that stops the light from H-alpha emission from being collected. It is possible to remove this filter and, at the time of writing, a Canon SLR camera with the filter removed is available from Hutech in the USA. With an appropriate white point setting, it can also be used as a daylight camera. Piggy-back AstrophotographyAfter experimenting with astrophotography with a fixed camera, either with just the camera itself using a fast film to allow short exposures, or using the camera body attached to a telescope to take photos of the Moon at prime focus or using eyepiece projection, a sensible progression would be to try piggy-back photography. This is the term applied when an equatorially mounted telescope is used to guide a camera mounted on the same equatorial platform. Camera lenses commonly used vary from 28 mm for wide-angle shots of the Milky Way to 300 mm for telephoto shots of large nebulae such as the North America nebula (NGC 7000) and the Eta Carinae nebula (NGC 3372). The basic requirements for piggy-back photography are:
Hand-operated slow motion controls on both axes could be used for lens focal lengths up to about 135 mm, but hand tracking in right ascension at the accuracy needed for a 300 mm lens will become tedious for exposures of more than about 10 minutes. Also, the operation of the controls could cause unacceptable, unintentional movements. The guiding tolerance at the plane of the film or CCD to ensure visually circular star images is 0.04 mm. (Gordon, 1985). Assuming a 135 mm lens is being used on the camera and the guidescope has a focal length of 1000 mm, the tolerable drift at the illuminated reticle of the guiding eyepiece is 0.04 x 1000 / 135, that is, 0.3 mm. Now, most guiding eyepieces have a 0.2 mm square box etched on the reticle, so the drift should be limited to 150% of the box size. Similarly, for a 50 mm lens, the drift can be up to 4 boxes and for a 300 mm lens, it must be limited to 65% of the box. (Different values for the tolerable drift will be obtained for different focal length guidescopes.) Film for piggy-back photography needs to be fine grained, as the image size is very small. Even with a 300 mm lens, the images of stars in galactic clusters will run together for all but the loosest clusters on the finest film. For colour prints, Kodak Royal Gold or Fuji Superia 800 are good choices as they have good red sensitivity, reasonably fine grain and can be used unhypered. Exposures of 10 to 20 minutes at f4 and 5 to 10 minutes at f2.8 will give acceptable results. For colour slides, Kodak Elite Chrome 200 (or the professional version, Ektachrome E 200) is the best choice due to its red sensitivity and its low reciprocity failure (that is, it maintains its speed throughout a long exposure). Exposures will be similar to those for the print films. Digital SLR cameras are also suitable, however most have built-in filters that exclude the red light of H-alpha. Digital images will need to include dark frames to enable electronic “noise” to be subtracted from the images, and even then, exposures will need to be relatively short with multiple exposures stacked if necessary. Advanced users will also use “flat fields” to enable them to remove the effects of dust on the detector and vignetting. A problem likely to be encountered is flexure. As the right ascension axis rotates, gravitational forces on the guidescope and camera vary with time and differential movement can occur unless everything is very rigid. For a 35 mm SLR camera with a 35 or 50 mm lens, a robust ball and socket bracket will be adequate, but for the same camera with a telephoto lens, a support for the lens as well as the camera will be necessary to reliably achieve untrailed star images.
Prime Focus AstrophotographyIn prime focus astrophotography, the telescope is used as a large telephoto lens. Prerequisites for prime-focus astrophotography are:
If using a digital SLR camera, it is recommended that the camera be powered from a large external battery to avoid the internal batteries dying during an exposure. Back To Start Of Prime Focus Section Prior to any photography session, it is necessary to spend time in planning. You should list all of the objects that you plan to photograph in the order that best suits their location in the sky. Many deep-sky objects will not be visible in the finderscope or on the focusing screen of the camera, so the image must be centred using star patterns taken from a good star atlas or a planetarium computer program. Copy or print out the star field and mark on it the field of view of the camera so that you can refer to it during the photography session. Estimate the time at the start of each exposure based on the planned length of the previous exposure and about ten minutes between exposures to allow for locating the object, finding and centring a guide star, resetting the timer and refocusing when necessary. Preplan the use of filters and the duration of all exposures. WRITE EVERYTHING DOWN. If, when your images have been printed, you cannot remember what you did, you won’t know why they turned out to be so good or so bad. Always have spare batteries on hand for any piece of equipment that relies on battery power. Photography without an illuminated reticle in the guiding eyepiece is impossible. The same goes for a battery-powered telescope drive. Back To Start Of Prime Focus Section An absolute requirement for prime-focus astrophotography is precise polar alignment. Imprecise polar alignment leads to frequent tracking corrections and frequent corrections will lead to poor guiding accuracy. It also results in field rotation if the misalignment is significant. If your telescope mount has an accurate polar alignment scope, then alignment will be readily achieved. If not, the star drift method should be used. The star drift method of polar alignment is both quick and easy (or reasonably so). If you can visualise the path in the sky followed by the telescope and the path of the stars for the telescope pointing towards the eastern horizon and pointing towards the celestial equator at the zenith, you wont need to memorise any rules, as corrections become obvious. If not, then remember that:
Keep making corrections until there is virtually no north-south drift in a five-minute period. To use this method it is necessary to know which direction is north or south in the field of view. One way of finding out is to nudge the telescope about its declination axis towards the south whilst looking through the eyepiece. Stars will appear to move north. Another method is to switch off the drive, which will establish west – the direction of drift of the stars. North is then 90 degrees clockwise from west for a Newtonian telescope that has two reflections of the starlight (or any even number). However, if there is an odd number of reflections then north is in the opposite direction. This occurs if you are using an off-axis guider with a Newtonian reflector or a Schmidt Cassegrain with a star diagonal. Back To Start Of Prime Focus Section Another requirement is comfort. Mosquitos can be a distraction so having access to insect repellent is useful and nights can be cold – even those that start out mild – so, apart from coastal locations in the summer months, warm clothing will be necessary. It is essential to be comfortable whilst guiding. Crouching or stretching to get your eye in line with the eyepiece will not result in the best-guided photo. Sitting in a good posture is ideal, however, if this is not possible, then standing erect is almost as good. Also, give your eye a break now and then, but not when the mount is at an inaccurate spot on the gears. By positioning the timer where it is easy to see, you will always get to look away from the eyepiece for a moment as you check on the number of minutes to go. (Time can seem to go very slowly at times.) Back To Start Of Prime Focus Section It is not possible to regularly achieve perfect focus using the focusing screen of an SLR camera, even if the regular screen is replaced with a fine ground glass screen. When you do focus perfectly, it is more due to good luck than to good management. However, for a film camera, a knife-edge focusing system can achieve near-perfect focus every time. How does such a system work? The light from a star directed on to a film, CCD or CMOS detector is in the form of a long, narrow cone, the apex of which is right at the plane of the photographic emulsion or detector when perfectly focused. Hence the image of a star will be a tiny point at the focus plane but a disc of light inside or outside the focus plane. If you can look down this cone of light from a position beyond the focus plane, you will see the disc of light. Now, whilst observing the disc of light, slide something thin and opaque (the knife) through the cone of light. If the knife is not at the focus, it will be seen as a shadow crossing the disc; inside focus it will appear to move in the same direction as the knife is moved and outside focus, in the opposite direction. However, at exact focus, the whole disc of light will be instantaneously obscured as the knife passes through the exact apex of the cone. At its simplest, for a film SLR camera, the knife is a piece of razor blade or blank exposed film glued to a piece of clear glass or acrylic of such a size that it can slide along the outer film guide rails in the back of the camera. The knife-edge should then be in the plane that the front of the film would be in when loaded in the camera. A piece of film or thin flat metal must be cut with a bevel to ensure that the front edge, not the back edge, cuts the light beam. The cutting edge of a piece of razor blade must be about 0.15 mm below the glass or acrylic carrier. A bevel-cut flat metal knife must also be this thickness. This works fine until a film is loaded into the camera. Beyond this point in time an alternative method is needed. One could purchase an identical camera body to use just for focusing, but it is not necessary to go to that expense when less than $10 spent at a hardware store will allow just as good a result to be achieved. Purchase the two parts of a 40 mm screwed PVC pipe coupling and trim one so that the coupled length is about the same as the depth from the front of the camera to the film plane. Some teflon tape wound around the threads make a tight fit and a piece of clear acrylic as a mount for a piece of razor blade is glued to one end. To adjust this simple tool, first focus the telescope-camera combination as previously described, then lock the telescope focuser and replace the camera with the PVC tool. By screwing it in or out, a setting can be reached which perfectly cuts the starlight beam at its focus. The PVC tool is then marked (it could be glued) and can thereafter be used to obtain good focus after a film has been loaded into the camera. Equivalent products are available for purchase from specialist manufacturers. (Check the advertisements in Sky & Telescope.) They can be locked into position, allowing the slow motion control on the mount to slowly move the knife edge across the light cone. With digital SLR cameras, there is no back that can be opened to allow the light cone to be viewed directly. There is therefore no option than to have a separate knife edge focusing device unless you are prepared to download an image to a computer and scrutinise the image at high magnification, repeating this until perfect focus is achieved. If you elect to make a knife edge focuser, you will need to get the dimension from the front of the camera to the detector from the camera manufacturer. This must be accurate to within about 0.05 mm or better. Always aim at a reasonably bright star at least 30 0 above the horizon (to avoid too much scintillation) and ensure that it is not a double star. Having achieved a good focus setting at the start of a session, you cannot assume that focus will be maintained over the next three or four hours of photography. At least two things may cause loss of focus. First is mirror shift (in a Newtonian or a Schmidt-Cassegrain telescope). Second is thermal expansion or contraction. If you focus at, say, 7 pm when the temperature is 18 0 and by 11 pm the temperature has dropped to 3 0, the telescope tube will have contracted enough to cause loss of focus, particularly if it is aluminium, and especially if it is a Schmidt-Cassegrain. The focus tolerance for an f6 system is about 0.07 mm – well able to be reached by thermal contraction or mirror shift. In summary, if you do not want your stars looking like blobs, or worse still, like doughnuts, then you must make the effort to achieve near-perfect focus and remember to refocus from time to time. Back To Start Of Prime Focus Section Along with focusing, guiding is essential to creating a good astrophotograph. People have been known to use hand-operated slow motion controls to guide during long exposures but it isn’t something you would want to do too often and the resulting photos would be the sort you would look at when nobody else is around! A good quality drive and drive corrector are essential, along with an illuminated reticle eyepiece. An autoguider is even better. Although it is preferable to have electric drives on both the right ascension and declination axes, hand operated slow motion control of the declination is acceptable if good polar alignment is achieved requiring a small correction only every four or five minutes. Great care must be exercised in making corrections, as it is easy to cause unintended movement in excess of that being corrected, particularly if the mount is not very sturdy. The number one enemy of good guiding is flexure – flexure of any part of the telescope and guiding system relative to the path followed by the light that falls on the film. During a 20-minute exposure, the right ascension axis rotates 5 0, which is enough to alter the gravitational forces acting on every part of the assembly. Movement could occur from flexure of the telescope tube, flexure of brackets holding a guidescope, flexure of the guidescope, flexure of a mirror mount, flexure of the rack and pinion focuser, flexure of the off-axis guider, etc. Flexure of brackets is probably the most common source of the problem. The best way to overcome flexure is to use an off-axis guider. This uses a small amount of the light, picked up and redirected sideways from the edge of the beam outside that which falls on the film or electronic detector. Advantages of this method are:
A disadvantage if you are using a Schmidt-Cassegrain telescope is that the guidestar images at the edge of the field are like blurred seagulls, not the crisp, round images you would like to see. Also, when photographing from the city where the low contrast between the bright sky and stars eliminates faint guidestars, finding a suitable guidestar will often dictate an off-centre image. Having composed the photo on the camera screen, found a suitable guidestar and adjusted the two axes of the telescope until the star is central in the guiding eyepiece, the task of guiding begins. (It is assumed that an autoguider is not being used.) Most illuminated reticle eyepieces have a 0.2 mm square box etched in the centre, often as two pairs of 0.2 mm spaced lines at right angles that cross the centre. Keeping the guidestar in that box is sufficient if photographing through a 135 mm telephoto camera lens but most certainly not sufficient for prime-focus photography through the telescope. It has been recommended that star images should not trail by more than 0.04 mm on the film. This can be represented as an amount of drift off line using the formula S=AF/57.3 where S is the image size (in this case the drift), and F is the focal length of the telescope. The number 57.3 is simply 180/pi, to convert radians to degrees. For example, for a 1500 mm focal length telescope, and after rearranging the formula, the drift A = 57.3 x 0.04 / 1500, that is, 0.0015 0 or 6 arc seconds approximately. Now the same formula can be applied to find the size in arc seconds of the 0.2 mm square box for the focal length of the telescope used. If a guidescope is being used, this will most likely be different from the photographic focal length. In this example, we will assume that an off-axis guider is being used, therefore the focal lengths will be the same. Thus, the box size A = 0.2 x 57.3 / 1500, that is 0.0076 0 or 28 arc seconds and our maximum drift is 20% of the box. Guiding within such a tight tolerance is made practicable by the magnification of the eyepiece. Guiding is necessary because no telescope drive is perfect – even the best observatory telescope. A good mass produced drive may have a tracking error of 20 arc seconds or more and a bad one, 120 arc seconds. The “best” tracking error is one that varies in a sinusoidal fashion; the worst is one that makes sudden periodic excursions. Always have the brightness of the illuminated reticle as low as possible in order that the star stands out clearly and so that, if the star goes behind one of the cross-hairs, you can detect the glow even if not the actual star image which, for a faint star, is tiny compared to the 0.02 mm or thereabouts width if the etched line. Do not guide using an out of focus star, as the small drifts that must be corrected are easy to detect only if the guidestar is well focused. There are many things that can go wrong. Most astrophotographers have, at some time or other, guided carefully for 20 or 30 minutes only to find that the camera shutter had been set at 1/125 second or a similar brief time! Clouds can be a nuisance when they move across unexpectedly, as can trees when the object being photographed sinks behind a branch before the exposure is complete. Aircraft, with cabin lights on and landing lights blazing, will ruin any photo if they cross the field of view. When things go wrong, cut your losses and start again. To begin an exposure, first hold the hand controller so that pressing the right hand right ascension control button moves the guidestar to the right, then position the sharply focused star in a corner of the box, start the timer, open the shutter and lock the cable release and you are on your way to creating what could be a magnificent photograph. Back To Start Of Prime Focus Section
All images by Max Kilmister. Back To Start Of Prime Focus Section Webcam ImagingIn the last few years imaging planets has become popular using commercially available web cameras. Philips released the popular ToUcam 740k a few years ago. A little over a year ago the ToUcam 840k was released which offered superior low light performance. Webcam imaging requires the acquisition of a movie sequence of a planet, which may contain hundreds or, thousands of frames. Webcam frames are of short duration and subsequently little detail is acquired in a single frame. A movie sequence is imported into a program such as Registax, which decomposes the movie into its individual frames. These frames are aligned, optimised and stacked by the Registax software. (Other image processing software such as AstroStack may be used.) The stacking process yields a frame which has a vastly improved signal to noise ratio over that of any individual frame in the movie. With the success of the ToUcam and other similar units in the hands of amateur astronomers, Meade soon launched the LPI (Lunar Planetary Imager) to be followed some months later by Celestron’s NexImage system. This article will describe the use of the Philips’ToUcam and Registax image processing software. Back To Start Of Webcam Imaging Section The Philips ToUcam 840K offers a maximum resolution of 640 x 480 pixels where each pixel is 5.6um x 5.6um. The camera is lightweight and provides excellent low light performance. The 840K is supplied with a screw in lens which is removed before imaging. A suitable screw-in adaptor is needed to connect the webcam to the telescope. The CCD pixel size must match the telescopes focal ratio to avoid undersampling. ‘(Undersampling is characterised by star images which appear blocky or square shaped.) Undersampling is avoided by choice of pixel size that provides sky coverage equivalent to half the “seeing” conditions or less. Backyard observing sites typically achieve “seeing” conditions of 3 to 5 arcseconds. The following formula defines the relationship between the three variables “seeing”, focal length and pixel size: (sampling in arcsecond)=206.265*(pixel size in microns)/(focal length in mm). Back To Start Of Webcam Imaging Section Philips provides a software bundle, VLounge with the camera. VRecord the image acquisition software is a component of the bundle. VRecord allows the user to capture an *.avi of the planet being imaged. VRecord includes controls to set the camera gain, brightness level and saturation. Back To Start Of Webcam Imaging Section A laptop computer is ideal, it offers mobility and battery operation. A Pentium 4 CPU is ideally suited to the task of aligning, optimising and stacking frames. The laptop must have a USB port for fast image data download from the camera. A USB extension cable is also typically required. Back To Start Of Webcam Imaging Section A planet’s image is captured as a movie sequence, a *.avi file. Given the camera is set for a high frame resolution and 10 to 20 frames per second are being acquired for some number of seconds the file size grows quickly. Consequently a large hard disk is required to save the file. A 20 gigabyte hard disk is the minimum with 40 gigabyte preferred. The image is downloaded to the computer over a USB connection. Given the large amount of data to be downloaded a fast CPU is desirable. To avoid “frame dropping” a Pentium 4 CPU is recommended. Back To Start Of Webcam Imaging Section One of the best available programs for processing the image sequence is Registax. Registax is a freeware program available to download from the Internet. Simply type Registax into a search engine such as Google to find a download site. The program allows a *.avi to be decomposed into a discrete number of frames. The user selects a frame from the *.avi which offers sharp focus captured at a moment of good “seeing”. This frame becomes the datum to which all other frames in the *.avi are referenced. The user now performs an alignment procedure. All frames in the *.avi are aligned to the datum and are ordered according to their “quality” with respect to the datum. The highest quality frames appear at the top of the order. The user now sets a quality cut-off point so only a certain percentage of frames are further processed. A cut-off point of 80% is normally sufficient so only 20% of frames in the *.avi progress to the next phase – optimisation. The optimisation phase attempts to improve the quality of the remaining 20% of frames. The stacking phase is next and allows the user to manually select how many frames they wish to stack by adjusting two slider controls. Having selected the desired frames the user presses the stack button. The stacking phase yields a single frame which has a vastly improved signal to noise ratio. Registax also includes quite powerful supplementary functions including unsharp mask to “sharpen” image features. If required the user can “touch up” the result using image processing programs such as Photoshop. Back To Start Of Webcam Imaging Section For imaging planets exposure times are typically 30 to 80 seconds. A telescope with at least an RA drive and reasonable tracking performance is ideal. The telescope does not have to be accurately polar-aligned. A telescope of focal ratio f9 and X2 or X3 Barlow gives an effective focal ratio of f18 or f27, which is fine for planetary imaging. This arrangement reduces the field size sufficiently to offer a good size planetary image. An aperture of 200mm affords adequate light gathering power. Back To Start Of Webcam Imaging Section Focusing is almost certainly the hardest aspect of CCD imaging. The Earth’s atmosphere is constantly changing. As a result planet image presented to the observer is going in and out of focus. On some evenings it is almost impossible to achieve good focus, particularly if the planet is low in the sky. Finding good focus is easier with a rack and pinion focuser than with a mirror shift focuser. An example focusing procedure is:
Back To Start Of Webcam Imaging Section That the above process yields outstanding results is demonstrated here. Figure 1 is an individual frame from a *.avi movie of the planet Jupiter. Note the poor quality of this frame, very little detail is apparent.
Registax processing of this *.avi yields Figure 2. Note the vast improvement in the quality of this image. Detail such as storm cells and equatorial banding is clearly visible.
Both images by Brent Joyce. There are number of tutorials on the web to help in using the Registax program. The AAQ also has a number of webcam users who can assist those interested starting webcam imaging. Back To Start Of Webcam Imaging Section
Back To Astrophotography Main Page Astronomical Association of Queensland 2006. www.aaq.org.au |
