Re: Canon 20 d

From: Greg Roberts (grr@telkomsa.net)
Date: Sat May 16 2009 - 22:31:32 UTC

  • Next message: GERHARD GROENEWALD: "Astrophotography."

    Evening Gerhard
    
    I think Scott has answered your questions so I wont try and repeat what he said. 
    For satellite work you do not need exposures longer than about 20-60 seconds on 
    the high altitude stuff - for leo satellites the maximum I use is 2.5 seconds- 
    longer than that and you may not get the whole trail in your field of view- 
    assuming you are working at a reasonable focal length of -at the very least 
    50mm - bit ideally around 100-200 mm - the longer your focal length the smaller 
    the field of view but the higher the positional accuracy since the number of arc 
    seconds per pixel  decreases- eg if scale 6 arc seconds per pixel then you may 
    be able to measure accurately to at least 1 pixel or, in other words 6 arc 
    seconds. For geostationary satellites I would recommend a focal length of 
    200-400mm - the longer the focal length the easier it is to pick up a 
    geostationary if you are trying to find it by picking up its motion relative to 
    the background stars.
    
    If you use a program such as APEX you can get sub pixel accuracy and are then 
    limited by other factors such as the "seeing" or steadiness of an image. I 
    typically have seeing around 12-16 arc seconds - terrible - so with an image 
    scale of say 6 arc seconds per pixel my star image will be "smeared" across 2-3 
    pixels - one of the reasons why its difficult to get a perfect focus. Now if I 
    went to the Sutherland Observatory the seeing can often get below 1 arc second 
    which means no smearing of the image and a deeper magnitude penetration since 
    all the starlight is concentrated in one pixel insteady of spread over 2 or 3.
    
    Whilst a camera or other optical system will normally be at its "fastest" - ie 
    most light sensitive - at its lowest f/number, so a lens at f/2.8 is "faster" 
    than a lens at f/3.5 and a lens at f/3.5 is faster than one at f/4.5, but 
    whether it will pay to use your fastest ( or smallest f/number) depends on your 
    local conditions and just how good the optical system is. If your sky is too 
    bright because of city lights etc you will only be able to do short exposures 
    before your image is "whited" out, so it sometimes pays then to "stop down" one 
    or two f/stops- you can then use a longer exposure and your background sky will 
    be darker. Even the best optical system -at the prices we can afford - produces 
    an optical image that is far from perfect when used at its smallest f/number. I 
    have a Nikon lens and at f/2.8 there is pronounced field distortion at the 
    edges - stopping down will improve the quality of the image.
    
    So the exposures you want are a few to several tens of seconds for high altitude 
    stuff, less than 2-3 seconds for leo sats-
    
    f/ratio - try your fastest but also try stopping down one or two stops and 
    compare image quality - you might even find you will go fainter if you stop down 
    one or two stops.
    
    To track at star or sidereal rate is not really necessary normally. For 
    geostationary or high altitude stuff it does not matter if the stars or slow 
    moving satellite produces trailed images. APEX was designed just for that kind 
    of situation. Since exposures for leo satellites are so short you probably wont 
    even notice how much the stars have trailed in an exposure 2-3 seconds long- if 
    you do its easy enough to know which end of the trail to measure .
    
    I sometimes use my CoSaTrak system to track on a satellite - this trails the 
    stars and the satellite appears as a stellar source-APEX can also handle this 
    kind of situation. I believe APEX can also handle stellar images obtained with a 
    sidereal drive- ie point sources - but I havent really explored that option yet.
    
    Okay - all this means is that you do not really need an equatorial mount with 
    drive to track on the stars but if you can afford it a GOTO mount does have 
    advantages in that you can tell it where to point and dont have to "star-hop" to 
    find a point in the sky. This is how I use my system - purely to swing to a 
    particular azimuth /elevation and wait for the satellite either to cross the 
    field of view, or if a slow moving object it should be in the field of view.
    
    Now as regards APEX. Scott and I have managed to get it to the point where it is 
    virtually 100% automatic. Originally is was a script driven program but that is 
    no longer the case. Because Scott and I have different equipment we have 
    approached the problem in two different ways. However the author of APEX will 
    now add this to APEX so it is done internally in the program and will not need a 
    simple program to prepare the script - which in my case takes about 2-5 seconds 
    for about 200 images. APEX is NOT difficult to use or learn - in fact its very 
    simple and I hope that once you get to the stage of producing useable images you 
    will make use of it. If you think HEAVENSAT is great then APEX is in the same 
    category . Ideally APEX does not like too long an exposure - we have found 
    around 5-20 seconds to be enough. If the exposure is too long star trails run 
    into each other and in a crowded star field, such as in the Milky way, APEX may 
    take some time to process the several hundred stars that may be visible....
    
    Incidentally what version of HEAVENSAT do you have - the latest is V2.17 - I 
    dont think its been put on the website for it yet.
    
    Thats enough from me - I am now "enjoying" my winter season which means little 
    or no optical tracking for the next 2-3 months so this gives me a chance to 
    upgrade/improve my system etc.
    
    Cheers
    Greg
    
    -------------------------------------------------------------------------
    Subscribe/Unsubscribe info, Frequently Asked Questions, SeeSat-L archive:  
    http://www.satobs.org/seesat/seesatindex.html
    



    This archive was generated by hypermail 2b29 : Sat May 16 2009 - 22:36:41 UTC