astrophoto. tips

Astrophotography

determining effective F ratio
determining effective focal length
calculating film image size
atmospheric extinction
determining correct exposure
photography problems
practical comet digital photography
see also:
- my astrophotography gallery index and more here: my pics
- astrophotography targets and lens/telescope sizes
- astronomic viewing - magnitudes, seeing, light pollution, techniques
- telescopes
- photography - basics
- using a digital camera for astrophotography
- CCD astrophotography
- advanced astrophotography and image processing
- Dan Heller's Tutorial Series- Photographing Star Trails / star trails in Tanzania
- Jim Solomon's cookbook for prime focus astrophotography
- photographing the moon:
  - my notes: the moon
  - Dan Heller's Tutorial Series- Photographing the Moon
- on the web:
  - Samir's guide for beginner astrophotographers
  - http://www.madpc.net/~firmament/astro/main_screen.html?hints_and_tips_dso_imaging.html&2
  - noise vs sub-exposures
  - Clark Vision's Canon 10D signal to noise tests
  - galleries of astrophotos:
    - by equipment used:
      - http://www.buytelescopes.com/gallery/browse_manf.asp
  - webcam photography:
    - Mogg's webcam adapter
    - Sylvain Weiller and his Vesta webcam mods to give longer exposures

Astrophotography:

see Astrophotography intro first for how to get started in astrophotography

General information and equations:

tip: if using film, always start a roll of film with a normal daylight or flash photo so that the film processors know where to start as astrophotographs can make their job very difficult!
tip: the best film cameras to use are those that can take long exposures without introducing light or vibrations, and for connecting to a telescope, it should be a SLR. A popular camera is the Olympus OM-1n as it has a mirror lock up and self timer to minimise vibrations and does not need batteries which often fail in cold weather.
tip: telephoto lenses for stars or comets will need guiding for exposures more than a few seconds. For comets, guide on the comet not stars as some comets move quite fast relative to the background stars (eg. 3/4 arcminute per minute).
tip: the best comet images are often taken using f/2 to f/2.8 rather than f/1.4, and avoid placing the comet head to near the edge where there tends to be aberrations.
tip: some lenses produce soft images as each star is surrounded by a unfocused blue halo. This can be reduced by using a light yellow filter such as Wratten 2B or 2E, but if using BW film, can use a stronger yellow filter.
tip: to accentuate a comet's blue gas tail, use a Wratten 47A filter, to accentuate the reddish dust tail, use a Wratten 21 filter;
determining the effective F ratio:
- prime focus method:
  - camera without lens, mounted to telescope without the eyepiece
  - f ratio = Barlow magnification x telescope objective focal length / telescope objective diameter
  - nb. if a Barlow lens is used, this method is referred to as negative projection method
  - see prime focus digital imaging
- afocal method:
  - see also: afocal photography explained
  - camera with lens, mounted to telescope with eyepiece
  - effective focal length (EFL) = Barlow magnification x telescope magnification x camera lens focal length
    - where, telescope magnification = telescope objective focal length / telescope eyepiece focal length
  - EFL = telescope objective focal length x projection magnification
    - where, projection magnification = Barlow magnification x camera lens focal length / telescope eyepiece focal length
  - f ratio = EFL / telescope objective diameter = f ratio of telescope x projection magnification
- positive projection method:
  - camera without lens, mounted to telescope with the eyepiece
  - if A = distance between the center of the eyepiece elements to the focus of the objective,
  - and B = distance between the center of the eyepiece elements to the film plane
  - then projection magnification = B/A, and,
  - f ratio = projection magnification x telescope objective focal length / telescope objective diameter
determining the effective focal length of system:
- effective focal length = effective f ratio x telescope objective diameter
calculating the film image size:
- for 35mm cameras:
  - film image size (mm) = angular diameter of subject in degrees * effective focal length of system (mm) / 57.3
  - film image size (mm) = angular diameter of subject in arc seconds * effective focal length of system (mm) / 202,265
  - thus, for 35mm film (36x24mm), the whole film will cover the following angular diameters:
    - 50mm lens (1x magnification) = 41.2° x 27.6°
    - 100mm lens (2x) = 20.6° x 13.8°
    - 200mm lens (4x) = 10.3° x 6.9°
    - 300mm lens (6x) = 6.9° x 4.6°
    - 900mm lens (18x) = 2.3° x 1.5°
    - 1200mm lens (24x) = 1.7° x 1.1°
    - 1800mm lens (36x) = 1.14° x 0.76°
    - 3600mm lens (72x) = 0.57° x 0.38°
    - 254mm aperture, 1200mm focal length telescope with 7.5mm eyepiece and 50mm camera lens:
      - f ratio = 160x mag. x 50/254 = f/31.5
      - effective focal length = 31.5 x 254 = 8000mm lens
      - angle of view = 0.26° x 0.17° = 928 x 619 arc seconds
- for Canon S30 digital camera:
  - 7mm wide angle (equiv. to 35mm lens for 35mm film) f/2.8: approx. 60° x 40°
  - 21mm zoom (equiv. to 105mm lens) f/4.9: approx. 20° x 13°
    - at 2048x1536 (3mpixels), moon would be ~ 50 pixels diameter (1/6th" on prints at 300dpi)
- for Olympus C8080 digital with teleconverter:
  - equiv. to 196mm focal length f/3.5 lens on 35mm camera
  - 3904 x 3417 pixels (8 mpixels) = ~11.8° x 10.4°
  - quarter moon:
    - ~ 165 pixels diameter - can resolve larger craters without a telescope (>1/2" on prints at 300dpi)
    - exposure 100ASA 1/250th sec (to see jupiter it is faint at 1/250th, so 1/15th sec brighter but then moon is over-exposed)
  - comet at magnitude 4: 400ASA head 40secs; tail 3-5min pending tracking, light pollution.
- binoculars:
  - 8x23 approximates 400mm f/17 camera lens - good for sports, concerts but not so good for astronomy
  - 7x50 means 7x magnification with aperture 50mm which approximates 350mm f/7 camera lens
- example approximate subject angular diameters:
  - comet with tail are often 5-10° long (ie. 200-300mm focal length is ideal), but faint tail may extend to 50°
  - nebulae are often 1-2° (ie. 900-1200mm focal length) eg. orion nebula = 1.1°
  - sun = 0.5333°
  - moon = 0.5116°
  - jupiter = 0.0111°
  - saturn = 0.0116°
  - venus = 0.0055°
  - mars = 0.0022° = 18-25 arc seconds at opposition (ie. approx. 100x smaller than the moon)
  - uranus = 4 arc seconds at opposition
what digital sensor for your telescope?
- there are several aspects that need considering:
  - image circle size projected by the telescope
    - image circle may not cover a 35mm full frame sensor
  - image quality of the periphery of the image circle:
    - you may need to use a smaller sensor size to avoid poor image quality in the periphery
    - alternatively, you may need to use a field flattener or reducer to improve image quality in the periphery
  - telescope/sensor resolution matching:
    - the goal of the telescope/camera combination is to optimise the amount of sky coverage each pixel sees with the object you are trying to image which in part is determined by sampling theory
    - approximation for focal lengths 200mm and greater:
      - arc-seconds/pixel = 206.265 x pixel size in microns / effective focal length in mm
        
        NB. 206.265 is derived from converting radians to degrees.
        
        general rule of thumb:
        
        for deep sky objects is to use a value 1.5 to 2.5 arc-secs/pixel
        
        for planetary objects, use 0.3-0.5 arc-secs/pixel for adequate resolution
        
        the higher this number is:
        
        the more sky will be covered by a camera with a given number of pixels
        
        the lower the resolution
        
        the higher the sensitivity to photons
        
        BUT if too large, stars will appear as blocks, thus may need to use a reducer to decrease the effective focal length.
        
        AND if too small, the image is said to be over-sampled as there are too many pixels per star which lowers sensitivity
        
        pixel size can be adjusted by "binning", thus a camera with 9 micron pixels when binned 2x2 will have effective pixel size of 18 microns, and if 3x3 will have 27 microns. Remember to decrease the pixel count for your FOV calculations though.
        
        alternatively, focal length can be increased as with Barlow or tele-converter
      - field of view of the camera = arc-seconds per pixel x number of pixels in that direction on the sensor
  - maximum acceptable length of sub-exposures:
    - depends on accuracy of tracking - quality and set up of mount
    - effective focal length - the greater this is, the more magnified are the tracking errors
    - light pollution - may limit sub-exposure times
    - light recording power of telescope system - aperture radius²/fratio²
    - sensor sensitivity (ISO)

atmospheric extinction:

for objects below 20deg altitude, multiply the film ASA by the ASA correction factor:

degrees altitude	extinction (stellar magnitudes)	ASA correction factor
0.75	8.78	0.00032
1	6.58	0.0024
1.5	4.39	0.018
2	3.29	0.049
5	1.32	0.298
7	0.94	0.423
10	0.66	0.546
15	0.44	0.671
20	0.33	0.739

determining the correct exposure:
- there is no exact method, thus one should bracket exposures
- for point sources of light:
  - telescope objective diameter is more important than f-ratio of focal length in determining correct exposure
  - a lens with twice the usable diameter will require half the exposure time
  - light recording power of a system = objective radius² / f ratio²
    - thus comparing a 300mm f/2.8 on a full frame camera with equivalent magnification as a 150mm f/2.8 lens on a Olympus dSLR, even though they would give the same magnification and f ratio, the former would have 4 x the light gathering power.
    - however, you can get a 150mm f/2.0 in the Olympus which has practically the same light gathering power as a 300mm f/2.8.
  - eg. omega centaurus:
    - 400ASA film 6" diam f/7 40min see pic
    - DSLR f/6 1.5 to 24min see pic 1 and 2
- for non-point sources of light:
  - use the f ratio as calculated above and the film ASA rating corrected for atmospheric extinction and reciprocity law failure (the following are for f/8 using corrected ASA of 500):
    - moon:
      - earthshine at 48hrs old: 4secs
      - 1st Q/ last Q: 1/125th sec
      - full moon( mag. -12.55): 1/500th sec
      - lunar eclipse: 1/4 in umbra: 1/500th sec; 3/4 in umbra: 1/250th; totality: 8secs;
    - sun (mag. -26.7; with no.4 neutral density filter):
      - 1/8000th sec
      - solar eclipse: 1/4 partial: 1/4000th sec; 3/4 partial: 1/2000th sec;
        
        no filters: diamond ring: 1/2000th sec; solar prominence: 1/4000th sec; corona: 1/250th
    - venus (mag. -4.07): 1/4000th sec (1/8000th sec at greatest brilliancy);
    - mars (mag. -1.85): 1/125th sec;
    - jupiter (mag.-2.2) : 1/1000th sec; 4 moons (mag. 4.6-5.6): 1 sec;
    - saturn (mag. 0.3-0.9): 1/60th sec; titan (mag. 8.4) 2 secs;
    - uranus (mag. 5.8): 4secs
    - ceres in opposition: f/2, 125ASA, 1 min;
- for deep sky objects such as nebulae: see deep sky astrophotography
photography problems:
- motion blur:
  - this may be caused by:
    - camera shake if shutter speeds < 1/focal length are used hand held
    - telescope vibration - esp. if windy, or camera mirror creates vibrations, consider using a self-timer
      - mild-mod. wind causes ~ 40-80 arc sec image shifts with a 10" Newtonian
    - rotation of earth:
      - the greater the magnification, the greater the effect of the rotation of the earth to cause "star trails"
      - a 50mm lens on 35mm film shows star trails if exposures are greater than 15secs
      - maximum exposure times are inversely proportion to effective focal length, thus 25mm lens can expose for 30secs
      - this can be minimised by tracking with either:
        
        dedicated camera equatorial tracking system without a telescope or auto-guide correction:
        
        Kenko Skymemo Tracker
        http://www.sciencecenter.net/hutech/kenko/skymemo/index.htm
        
        It can be attached to a photographic mount, so there is very little extra weight in addition to the ordinary photographic gear required.
        
        Vixen Photoguider
        http://www.teleskop-service.de/Vixen/Astrofotografie/astrofotografie.htm#PhotoGuider
        
        this is, in fact, a GP mount but without the declination part, and with a platform
        for fixing two cameras. Also, the mount is more lightweight.
        
        manually tracking:
        
        guiding an object by viewing through a telescope during the exposure when there is no RA motor drive on the telescope mount
        
        auto-tracking:
        
        use of a motor drive to drive the telescope's equatorial mount in sync with earth's rotation
        
        on most amateur mounts, even if accurately balanced and aligned, there is a periodic error every 5min with peak-to-peak variation of between 15-120 arc secs, the best mounts have errors of as low as 2 arc secs.
        
        auto-guiding:
        
        for best results, one needs to use an auto-guider that elecronically adjusts both the RA & Dec drives of the mount in order to keep the image of the star in a constant position, this requires a mount capable of this adjustments (eg. LX200, LXD75, Losmandy but not HEQ5 or LXD55) and a CCD sensor to detect the star. The various options are:
        
        off-axis CCD sensor options:
        
        CCD on separate guide scope
        
        These give the greatest flexibility in choosing a guidestar, and filters on the main camera have no effect on the guide image. The downside, is that it cannot correct for either mirror shift in the main scope, or deflection of any part between the two scopes. However if imaging durations are kept reasonable (perhaps five minute exposures), and reasonable care is taken, this approach is the most flexible one. Remember also the weight penalty for the second scope
        
        the guide scope could be a cheap 60mm f/15 refractor
        
        CCD on off-axis guider attachment - requires sufficient back-focus of the scope to allow it - may be a problem with Newtonians.
        
        CCD camera with inbuilt off-axis CCD sensor (eg. SBIG)
        
        These remove part of the 'unused' light, outside the normal imaging area. The SBIG cameras do this, as do things like the Lumicon GEG, the Van-Slyke guider etc.. Advantage is it retains the ability
        to correct for movements in the optical system, and costs nothing in
        sensitivity. The 'downside', is that the selectable guide area, normally
        has a small range of movement relative to the main CCD (for the SBIG system, none, unless you turn the whole camera), limiting the choice of guide stars. In the case of the SBIG system, filters also filter the light path to the guider, which can make guiding harder when using narrowband filters.
        
        on-axis CCD sensor:
        
        use a partially reflective mirror or prism in the main light axis giving best guiding as you can guide on the target and adjust for changes i movement of the optical system but decreases light to the camera
        
        eg. Star2000
- vignetting:
  - this is when the edges are blackened without image
  - prevent this by ensuring the camera lens diaphragm is positioned at the point of focus of the image
    - this may be impossible if using modern cameras with large zoom lens which have the diaphragm quite proximal - try using 2" eyepieces with long eye relief
  - alternatively one can use the SLR camera to determine how close the camera must be to the telescope for optimum image coverage
  - new eyepieces such as the MaxView 40 (1.25") and MaxView II (2") are designed to reduce vignetting with digital cameras
  - minimise vignetting by:
    - Use an eyepiece with long eye relief.
    - Couple the end of the camera lens as close as possible to the eye lens of the telescope eyepiece.
    - Set the digital camera at macro mode rather than infinity.
    - Use digital camera at full optical zoom.
    - Purchase lenses specifically designed for the type of digital camera used.
    - If possible, use a camera lens with a focal length longer than the eyepiece focal length.
- excessive contrast:
  - very bright comets (rare as they are) can be difficult to photograph the bright coma as well as the dim tail
- reciprocity law failure:
  - films tend to become less efficient at long exposures (minutes) and thus corrections need to be made for this effect
  - normally, the same exposure can be obtained by doubling the exposure when the f ratio is altered by one f-stop, but as exposures exceed a certain duration for a given film, this relationship fails, thus one may need to quadruple the exposure time when the exposure time is 30 secs and wish to change f ratio by 1 f-stop!
    - in general, to compensate, add 1 f-stop for 1sec, 2 f-stops for 30sec & 3 f-stops for 120 secs
  - for a great pic of stars in a dark sky, the following are approx. equivalent exposures on an f/2 lens:
    - 15 minutes with 400ASA = 30-60secs with 1000ASA film!
    - therefore try using fast film such as:
      - color slide film:
        
        Kodak Ektachrome P1600 and Fuji Provia 1600 have base speed of 400ASA but can be push-processed to 3200ASA
      - color negative film:
        
        Kodak Royal Gold 1000; Fuji Super G 800; Konica SRG 3200;
      - black and white film:
  - see film characteristics for degree of failure.
  - fortunately this is not a problem with digital photography