Re: Optical 30 Nov 2012

From: Marco Langbroek (marco.langbroek@online.nl)
Date: Sat Dec 01 2012 - 16:35:56 UTC

  • Next message: Kevin Fetter: "Pleiades 1B launch in a few hours"

    Op 1-12-2012 15:55, Greg Roberts schreef:
     > (2) Whilst not as good as my ccd camera-which costs a great deal
     >      MORE - some useful work could be done with this and similar
     >      DSLR cameras. Much experimenting would be required as one can
     >      manually set such items as aperture, f/ratio, ISO value (this
     >      is the "sensitivity"), size of image in pixels - the more
     >      pixels the better the image scale and hence resolution and
     >      other items not so important like white balance, noise
     >      reduction etc
     >
     > (3) The main problem is timing accuracy. This particular camera
     >      gives the time the image was taken to the nearest 0.1 second
     >      using its internal clock which the user must set. At start of
     >      the observing session I set the camera clock to the nearest
     >      second against my GPS digital display to 0.xx seconds.
     >      I then photographed the display and this gave the "time offset
     >      error".
     >      At the end of the observing session I again photographed the
     >      GPS display and got the new time offset error. This showed
     >      that the camera clock drifted +0.3 seconds in 195 minutes.
     >      I made the assumption that the clock error is linear but for
     >      high altitude satellites were a timing accuracy of 1 second
     >      is usually adequate this is not a problem but I applied this
     >      correction to all the times determined.
    
    
    I work exclusively with a DSLR for tracking. If you know what you are doing, 
    they are very good instruments for tracking.
    
    Calibration is necessary: not only the shutter delay, but also for the fact that 
    a "15 second" exposure is not really 15.00 seconds (in fact, on Canon EOS 
    cameras "15 seconds" is closer to 16 seconds in reality).
    
    I currently use a Canon EOS 60D DSLR. I do not use the internal clock at all. I 
    manually try to trigger the camera (with electronic wire release of course) as 
    best as possible at a whole .00 second, looking at the display of a DCF77 
    radiocontrolled clock. Then I apply a time correction previously obtained by 
    calibrating my images on Iridium satellites and satellites simultaniously 
    photographed and video-imaged (video with GPS time inserter). After quite some 
    practise, I now generaly have times accurate to better than 0.1s.
    
    The good thing about DSLR's is that a vast array of fast photographic lenses is 
    available, making your gear very versatile.
    
    Another nice thing about DSLR's is that they provide you with a large FOV, 
    larger than with a typical CCD camera (or video). You are also less dependant on 
    access to a power source, and your gear is very easy to take with you.
    
    For GEO objects I use a SamYang 1.4/85mm lens which performs marvelously. For 
    LEO I use (depending on the object) a Canon EF 2.5/50mm, an EF 2.0/35mm or the 
    already mentioned SamYang 1.4/85mm. For HEO I mainly use a 2.8/180mm Zeiss.
    
    I mainly use 10 second (in reality 10.05 second) exposures at ISO 1000 (or 1600 
    near the zenith if targetting HEO's), working from the midst of a town.
    
    Hope this is valuable info for anyone considering employing a DSLR for tracking 
    him/herself. There is a bit of a learning curve to it, but with some effort they 
    give you quite accurate results.
    
    - Marco
    
    
    -----
    Dr Marco Langbroek  -  SatTrackCam Leiden, the Netherlands.
    e-mail: sattrackcam@langbroek.org
    
    Cospar 4353 (Leiden):   52.15412 N, 4.49081 E (WGS84), +0 m ASL
    Cospar 4354 (De Wilck): 52.11685 N, 4.56016 E (WGS84), -2 m ASL
    Station (b)log: http://sattrackcam.blogspot.com
    Twitter: @Marco_Langbroek
    -----
    _______________________________________________
    Seesat-l mailing list
    http://mailman.satobs.org/mailman/listinfo/seesat-l
    



    This archive was generated by hypermail 2b29 : Sat Dec 01 2012 - 16:36:52 UTC