Re: CCD satellite tracking

From: Richard Crisp (
Date: Thu Feb 21 2002 - 15:00:33 EST

  • Next message: Matson, Robert: "SkyMap screen capture"

    Hi folks,
    I usually lurk this list looking for intersting tips of which I have found
    Some of you know that I am really a CCD astronomer, but we imagers often
    find that we capture your birds in our images.
    Unlike many that feel disappointed when a satellite track "spoils" their
    image, I feel lucky because I get to do some detective work.
    From the help of several on this list including Bjorn Gimle and Edward S.
    Light, I can now run SkyMap and figure out what bird I captured.
    What fun!
    Since most of my astro-images are composed of multiple short exposures to
    prevent saturating the detector and having to deal with blooming, I use
    median combine in my stacking of the image elements. By using median
    combining rather than averaging or summing the images, a satellite track
    magically disappears from the final image. So rather than getting angry I
    truly get a "twofer": my originally planned image and the fun detective work
    I get to do afterwards.
    Most of the time when I image I use the autosave feature of my CCD software
    and that ensures that  the date stamp represents the end time of the image
    capture. So knowing the astronomical object, the date and the time of the
    image, I have everything I need to figure out what made the track provided
    the object is in the TLE set.
    You can see one of my images that contain tracks at:
    Unfortunately as I write this (noon Pacific Standard Time in the USA), my
    webpage server is down. So if it doesn't work for you now, come back later.
    ----- Original Message -----
    From: "Greg Roberts" <>
    To: "Seesat" <>
    Sent: Thursday, February 21, 2002 5:12 AM
    Subject: CCD satellite tracking
    > Hi
    > This was prepared for another satellite group but since this might
    > be of interest to others I am posting it also to SeeSat
    > CCD Imaging:
    > ~~~~~~~~~~~
    > As some of you know I use a CCD imaging system that is proving fairly
    > successful and I would like to encourage others to do something similar
    > as it makes observing more comfortable and one can end up with a recording
    > of an observing session that one can analyse later.
    > There are currently two ccd cameras available that look like they have
    > potential - this is the PC164C and the 2006A. The former is available
    > from
    > for $129 and has a quoted sensitivity of 0.0003 lux and the latter is
    > available from
    > model 2006X at 69 UK pounds and sensitivity 0.003 lux- see also
    > model 2005XA.
    > I DONT have either but they appear suitable and are more sensitive
    > than my 0.01 lux ccd. (Standard disclaimer -- I dont have any
    > connection with this company blah blah blah - except that I cannot say
    > I am a satisfied customer since I do not have either ( but Im seriously
    > thinking about it!))
    > Okay, assuming you are prepared to waste some money, where does one start
    > and what does one need?.
    > First of all these are video cameras so, as they stand, you cannot capture
    > still images (except perhaps as a screen capture) nor do long exposures (
    > little use in satellite tracking unless you want a satellite trail).  The
    > shutter speed is automatically determined by lighting conditions and you
    > normally have no control over such . Depending on where you live in the
    > world you either need an NTSC or a PAL version - as far as I can see the
    > PC164C is an NTSC version ONLY and the 2006A is a PAL version - make sure
    > you get the correct version otherwise you may have problems displaying
    > images from such. The version to get will be the same (NTSC or PAL) that
    > your television set uses.
    > The cameras supplied do NOT come with lenses or a power supply. I suggest
    > you get the appropiate 12v DC power supply when you purchase the unit. Do
    > NOT buy a lens - these are expensive and have little use in satellite
    > tracking. However if you are feeling flush with money then go ahead - at
    > least you will be able to make videos of your family etc!
    > The units probably have a BNC connector on the back for your video output
    > and a socket for the 12v power supply- sometimes these are both in the
    > cable -ie one connection and the cable carries both video and +12v DC. You
    > can have a fairly long run of co-axial cable - in my case my CCD camera
    > with a 30 metre length - anyway please make sure you get the correct cable
    > and power supply when ordering the particular camera.
    > How does one display the image coming out of the coax cable ?  If you have
    > VHS video recorder this is the easiest way-
    > plug the video coax cable into the video input of your recorder. (You may
    > have to switch your recorder into AV mode) . All being well, with the
    > recorder switched on, you can record the video signal and at the same time
    > view the image on your TV set - assuming you have connected your video
    > recorder so you can watch videos on your TV set. This is probably the
    > cheapest way to do it - if you have a video recorder that is!
    > Alternatively you need a composite video monitor - if you played around
    > the early Apple and TRS80 computers (and probably others BBC ? /Atom/ZX81)
    > you will have this lying around somewhere. Mine come from a closed circuit
    > TV system used in a local bank where they had a security monitoring system
    > installed.  Buying these monitors new is quite expensive but you should be
    > able to pick up one quite cheaply in a second hand computer shop/flea
    > etc -  again make sure that the monitor is video composite and either NTSC
    > or PAL and working  ( as you need).
    > Another alternative exists and that is to buy a TV card for your
    computer -
    > these normally have a composite video input socket and it allows you to
    > the image on your computer monitor. It is however not very practical to
    > store the video data on your pc as the files will be ENORMOUS.  I record
    > onto a video recoder - the standard domestic VHS recorder - and then play
    > the tape back into my TV card in the computer. The TV cards are quite
    > cheap - you dont need anything very fancy-just make sure you have a
    > composite video input socket. Try and avoid a USB card as this only adds
    > unneccesary expense.
    > The only other thing you need is a lens. These I buy second hand from used
    > camera dealers or shops that buy someone elses junk cheaply and sells to
    > at a reasonable (sometimes!) profit. Such lenses are remarkably cheap
    > invariably they are obsolete ( in terms of current cameras) and have odd
    > connections to a camera - eg you cannot fix a Leica lens onto a Minolta
    > camera body. In our case we dont have to bother - we will probably have to
    > devise our own way to attach the lens to the ccd body so normally can
    > the connection mechanism.
    > For satellite use we are looking at getting a fairly decent magnitude and
    > the same time getting a reasonable field of view.
    > This limits us somewhat as its difficult to get a 3 degree field on a
    > reasonable sized aperture, so look at getting a lens with a focal length
    > around 70 to 105 mm - I use a 25mm focal length lens on my wide angle
    ccd -
    > field about 10 degrees, and a 70 mm on my narrow field ccd - field about
    > degrees.  Go for as fast an f/ratio as possible.... unfortunately its
    > difficult to get a 70mm lens at f/1.2  and a typical 105mm lens is only
    > f/3.5. The f/ratio controls how bright the image is, and thus how faint
    > can see so if you had to choose between a f/1.8 lens and an f/2.8 lens of
    > the same focal length then go for the f/1.8.
    > Remember f/ratio controls brightness of image and how faint you can "see"
    > and focal length controls your field of view.
    > There is no need to have a fancy tracking system like CoSaTrak - all you
    > need is a crude alt/azimuth mount that can point accurately to about 1
    > degree and you point it at the appropiate region of sky . Make sure the
    > is focused properly - focus is extremely sharp since we are working with
    > fast lenses at short focal lengths - it should focus on infinity-assuming
    > you have been able to get the lens to focus properly onto the ccd chip
    > your homemade lens/ccd body attachment. Plug the video signal cable into
    > your recoder/video monitor/computer and sit down in comfort in a warm room
    > and watch the screen.
    > You should see stars - adjust the monitor controls to MAXIMUM contrast and
    > your BRIGHTNESS control to give a background that is grey -- ie you can
    > actually see the speckles of the ccd electron noise. If you turn the
    > brightness down to show the stars as white dots on a black background then
    > you will miss the fainter satellites, so turn up your BRIGHTNESS so the
    > background noise is clearly visible.
    > You should now be able to see satellites cross the field. If you want to
    > further then things get a little more complicated since you need accurate
    > time somehow - I have a homemade time display unit which displays the time
    > on the screen so its also recorded on video tape. If you video record the
    > video coming in then you have to devise a way to identify the star field
    > make satellite measurements - that could be the subject of another "HOW
    > TO..." if anybody wants it. The purpose of this HOW TO is just to be able
    > see satellites.
    > Finally what sort of magnitudes can one hope to see-- I live virtually in
    > the middle of Cape Town so I have terribly bright skies - so much so I
    > considered it a waste of time trying to see satellites naked eye - the
    > eye limit being usually around +3.5 and maybe +4.5 on a good night-
    > depending in which direction I look. With my 0.01 lux camera I can see
    > down to about magnitude +8.5, and satellites down to about +7.8.  At a
    > guess I would say the 0.003 lux camera will get you about a magnitude
    > fainter, and the 0.0003 lux camera even more - its difficult to give more
    > accurate figures . If you have darker skies you should see even fainter -
    > think I am safe in saying that whatever magnitude you can see in 7x50
    > binoculars you will see with these CCD sensitivities.
    > For those that use a fixed telescope/binoculars for observing you could
    > attach the ccd camera onto the binoculars/telescope (similar to a
    > and have a video recorder running all the time whilst you observe visually
    > - you will then have a record of what you looked at and in many cases will
    > have the satellite image to look at afterwards and confirm your
    > observations.
    > Just make sure you use a long enough video tape.. eg a 240 minute tape if
    > observing for more than 3 hours . I dont recommend using the slow speed
    > on a video recorder - the image quality is bad enough without making it
    > worse!
    > I think I have covered all the main points - if anyone has questions Ill
    > pleased to answer if possible.
    > Cheers
    > Greg
    > -----------------------------------------------------------------
    > Unsubscribe from SeeSat-L by sending a message with 'unsubscribe'
    > in the SUBJECT to

    This archive was generated by hypermail 2b29 : Thu Feb 21 2002 - 14:59:26 EST