Loading ...
Sorry, an error occurred while loading the content.

Re: Exposure time for luminescence images

Expand Messages
  • Stan
    ... It depends on what you mean by exposure time : the length of a sub-exp or the total exp time for a target/image? There is really no difference in regards
    Message 1 of 8 , Feb 10, 2013
    • 0 Attachment
      --- Mark Striebeck wrote:
      > I found some interesting articles about how to evaluate the
      > different exposure times for the RGB images. But I couldn't
      > find anything how to determine the exposure time for the
      > luminescence images...

      It depends on what you mean by "exposure time":
      the length of a sub-exp or the total exp time for a target/image?

      There is really no difference in regards to sub-exp S/N dynamics. Read noise is constant regardless of filter and so the sky background level necessary to bury the read noise remains the same regardless of filter, f-ratio and such. However, filtered exposures (esp with slower f-ratio and/or dark sky) may not be able to achieve "sky limited" within a reasonable time, which is one reason for binning the RGB in an LRGB (2x2 binned pixels collect 4x more photons per same amount of readnoise).

      Total exp time is a wholly different subject and is not much explored in the amateur community. Aside from the important issues of image depth and definition (MTF, limiting mag, pixel and object S/Ns) there is the question of proportional time allocation over the 4 different pass-bands. The RGB time allocations are fairly straighforward but the L/(RGB) ratio is complicated by divergent multiple outcomes: depth, resolution, color definition and color fidelity. I've not seen a rigorous treatment of the full dynamics (though I haven't looked very hard). My informal average of reported ratios is approx L/(RGB) = 3/(1:1:1), which may represent a stochastic convergence for typical pretty pix imaging.

      Stan
    • Mark Striebeck
      Sorry, should have been more specific. What I m interested in is the relative exposure time (sub, but also overall) of luminance vs. rgb. MarkS ... [Non-text
      Message 2 of 8 , Feb 10, 2013
      • 0 Attachment
        Sorry, should have been more specific. What I'm interested in is the
        relative exposure time (sub, but also overall) of luminance vs. rgb.

        MarkS


        On Sun, Feb 10, 2013 at 9:40 AM, Stan <stan_ccd@...> wrote:

        > **
        >
        >
        > --- Mark Striebeck wrote:
        > > I found some interesting articles about how to evaluate the
        > > different exposure times for the RGB images. But I couldn't
        > > find anything how to determine the exposure time for the
        > > luminescence images...
        >
        > It depends on what you mean by "exposure time":
        > the length of a sub-exp or the total exp time for a target/image?
        >
        > There is really no difference in regards to sub-exp S/N dynamics. Read
        > noise is constant regardless of filter and so the sky background level
        > necessary to bury the read noise remains the same regardless of filter,
        > f-ratio and such. However, filtered exposures (esp with slower f-ratio
        > and/or dark sky) may not be able to achieve "sky limited" within a
        > reasonable time, which is one reason for binning the RGB in an LRGB (2x2
        > binned pixels collect 4x more photons per same amount of readnoise).
        >
        > Total exp time is a wholly different subject and is not much explored in
        > the amateur community. Aside from the important issues of image depth and
        > definition (MTF, limiting mag, pixel and object S/Ns) there is the question
        > of proportional time allocation over the 4 different pass-bands. The RGB
        > time allocations are fairly straighforward but the L/(RGB) ratio is
        > complicated by divergent multiple outcomes: depth, resolution, color
        > definition and color fidelity. I've not seen a rigorous treatment of the
        > full dynamics (though I haven't looked very hard). My informal average of
        > reported ratios is approx L/(RGB) = 3/(1:1:1), which may represent a
        > stochastic convergence for typical pretty pix imaging.
        >
        > Stan
        >
        >
        >


        [Non-text portions of this message have been removed]
      • tpiccian
        I try to get ADU counts of the brightest areas in the 30-40k (80% of full well capacity) range. I do a test exposure to measure brightness. If the backround
        Message 3 of 8 , Feb 11, 2013
        • 0 Attachment
          I try to get ADU counts of the brightest areas in the 30-40k (80% of full well capacity) range. I do a test exposure to measure brightness. If the backround sky doesn't cooperate I reduce the exposure time. Usually in my prime site, Cherry Springs State Park, that amounts to roughly 10 minute exposures on my SCT at F/6.2. Much will depend on the object.

          I average all stacked images, using Adam Block's suggestions in his CCDStack and Photoshop lessons. Adding pixels results in ADU counts that exceed the 65k values. I've been told that doesn't help the image but I'm not sure why.

          He also has a very good method for dealing with bright nebulosity in images like M42 and M20. Essentially he combines a short exposure image that shows detail in the brighter areas with a longer exposure that shows the detail in the dimmer areas. And it worked surprisingly well.

          Tom P.



          --- In ccd-newastro@yahoogroups.com, Mark Striebeck wrote:
          >
          > Sorry, should have been more specific. What I'm interested in is the
          > relative exposure time (sub, but also overall) of luminance vs. rgb.
          >
          > MarkS
          >
          >
          > On Sun, Feb 10, 2013 at 9:40 AM, Stan wrote:
          >
          > > **
          > >
          > >
          > > --- Mark Striebeck wrote:
          > > > I found some interesting articles about how to evaluate the
          > > > different exposure times for the RGB images. But I couldn't
          > > > find anything how to determine the exposure time for the
          > > > luminescence images...
          > >
          > > It depends on what you mean by "exposure time":
          > > the length of a sub-exp or the total exp time for a target/image?
          > >
          > > There is really no difference in regards to sub-exp S/N dynamics. Read
          > > noise is constant regardless of filter and so the sky background level
          > > necessary to bury the read noise remains the same regardless of filter,
          > > f-ratio and such. However, filtered exposures (esp with slower f-ratio
          > > and/or dark sky) may not be able to achieve "sky limited" within a
          > > reasonable time, which is one reason for binning the RGB in an LRGB (2x2
          > > binned pixels collect 4x more photons per same amount of readnoise).
          > >
          > > Total exp time is a wholly different subject and is not much explored in
          > > the amateur community. Aside from the important issues of image depth and
          > > definition (MTF, limiting mag, pixel and object S/Ns) there is the question
          > > of proportional time allocation over the 4 different pass-bands. The RGB
          > > time allocations are fairly straighforward but the L/(RGB) ratio is
          > > complicated by divergent multiple outcomes: depth, resolution, color
          > > definition and color fidelity. I've not seen a rigorous treatment of the
          > > full dynamics (though I haven't looked very hard). My informal average of
          > > reported ratios is approx L/(RGB) = 3/(1:1:1), which may represent a
          > > stochastic convergence for typical pretty pix imaging.
          > >
          > > Stan
          > >
          > >
          > >
          >
          >
          > [Non-text portions of this message have been removed]
          >
        • Stan
          ... Lum sub-exp time should be determined by its own criteria (e.g. sky limited ) and has nothing to do with the sub-exp times of the RGB. Any sub-exp L/RGB
          Message 4 of 8 , Feb 11, 2013
          • 0 Attachment
            --- Mark Striebeck wrote:
            > What I'm interested in is the relative exposure time
            > (sub, but also overall) of luminance vs. rgb.

            Lum sub-exp time should be determined by its own criteria (e.g. "sky limited") and has nothing to do with the sub-exp times of the RGB. Any sub-exp L/RGB ratio is happenstance and irrelevant (though once you settle into an imaging regime, that ratio will become somewhat constant).

            The L/(RGB) ratio of TOTAL exp times is important but as the prior msg stated, the dynamics are not quantitatively well understood and so most people experiment to find a ratio that works for them. That ratio can often be a result of circumstance; for true resolution it is highly advantageous to shot Lum when the seeing is good and fill-in the color data when the seeing is otherwise not useful.

            Basically the ratio of total Lum time to total RGB time determines the balance of color fidelity vs. image definition/depth. Pure RGB produces the highest color fidelity and color-definition but also produces poor depth and resolution. Pure Lum (otherwise known as grayscale or B&W) produces the greatest depth, contrast and resolution. A strong Lum with under-done coloration data can produce washed-out/undifferentiated colors.

            Stan
          • Mark Striebeck
            Thank you for the explanation - makes sense. MarkS ... [Non-text portions of this message have been removed]
            Message 5 of 8 , Feb 11, 2013
            • 0 Attachment
              Thank you for the explanation - makes sense.

              MarkS
              On Feb 11, 2013 11:19 AM, "Stan" <stan_ccd@...> wrote:

              > **
              >
              >
              > --- Mark Striebeck wrote:
              > > What I'm interested in is the relative exposure time
              > > (sub, but also overall) of luminance vs. rgb.
              >
              > Lum sub-exp time should be determined by its own criteria (e.g. "sky
              > limited") and has nothing to do with the sub-exp times of the RGB. Any
              > sub-exp L/RGB ratio is happenstance and irrelevant (though once you settle
              > into an imaging regime, that ratio will become somewhat constant).
              >
              > The L/(RGB) ratio of TOTAL exp times is important but as the prior msg
              > stated, the dynamics are not quantitatively well understood and so most
              > people experiment to find a ratio that works for them. That ratio can often
              > be a result of circumstance; for true resolution it is highly advantageous
              > to shot Lum when the seeing is good and fill-in the color data when the
              > seeing is otherwise not useful.
              >
              > Basically the ratio of total Lum time to total RGB time determines the
              > balance of color fidelity vs. image definition/depth. Pure RGB produces the
              > highest color fidelity and color-definition but also produces poor depth
              > and resolution. Pure Lum (otherwise known as grayscale or B&W) produces the
              > greatest depth, contrast and resolution. A strong Lum with under-done
              > coloration data can produce washed-out/undifferentiated colors.
              >
              > Stan
              >
              >
              >


              [Non-text portions of this message have been removed]
            • hilary15xx
              Hi Mark, You might want to try my SimCCD and Exposure Planner utilities. They can be used to give a recommended exposure time for that part of the image that
              Message 6 of 8 , Feb 15, 2013
              • 0 Attachment
                Hi Mark,

                You might want to try my SimCCD and Exposure Planner utilities. They can be used to give a recommended exposure time for that part of the image that is most important to you. For example if you want to get a good signal to noise ratio in the tidal tail of a galaxy, you can select that region and these utilities will tell you how many subs to take. You can find more information about these utilities at
                http://www.darklights.org/simccd/index.shtml

                Hilary

                --- In ccd-newastro@yahoogroups.com, Mark Striebeck <mark.striebeck@...> wrote:
                >
                > Hi,
                >
                > I found some interesting articles about how to evaluate the different
                > exposure times for the RGB images. But I couldn't find anything how to
                > determine the exposure time for the luminescence images. Is there a general
                > rule or are there good sources to read up on it?
                >
                > Thanks
                > MarkS
                >
                >
                > [Non-text portions of this message have been removed]
                >
              Your message has been successfully submitted and would be delivered to recipients shortly.