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

looking for camera (or workflow) recommendation

Expand Messages
  • Orlando Andico
    this is a bit related to my queries yesterday. Here s my situation: I live in an extremely light-polluted location. I accept that AP under these conditions is
    Message 1 of 29 , Mar 14, 2013
    • 0 Attachment
      this is a bit related to my queries yesterday.

      Here's my situation: I live in an extremely light-polluted location. I
      accept that AP under these conditions is severely compromised, but I
      would like to maximize what I can do.

      Right now I'm using a... QHY8 OSC. And for filtration, I'm using a
      stacked IDAS LPS-V4 and Red #25. This is broadly equivalent to a 20-nm
      bandpass H-a filter.

      What, you say, OSC?! yes only 1/4 of the pixels get any light. The
      QHY8 is rated at 50% QE in H-a (at least, for the pixels that actually
      get any light) and has 7.8um pixels. Read noise is 10e- and dark noise
      is... very small. Probably not worth considering, since it's in the
      same ballpark as the ICX285 and ICX694.

      With a 10-minute exposure with the above setup on the Rosette, my
      signal is at about 2400 - 4300 ADU. That is only a 2.5dB dynamic
      range. It is well separated from the left side of the histogram.

      With a 20-minute exposure, the signal is at about 3700 - 7500 ADU,
      which is a 3dB dynamic range. Still pretty bad, and at 20 minutes I
      run into field rotation issues between subs.

      So.. I am no expert, which is why I'm here.

      1) it looks like even 10-minute subs are already more than enough,
      because the signal is well above the pedestal. True or false?

      2) that dynamic range (this is the Rosette) is pretty darn small. How
      can I get it higher? - is this a camera issue, a narrower-filter
      issue, or... my skies just well and truly suck?

      10-minute sub - https://dl.dropbox.com/u/63497702/Rosette_1_600sHa_00003.FIT

      20-minute sub - https://dl.dropbox.com/u/63497702/Rosette_1_1200sHa_00000.FIT



      Question is - what do I need to do to get better results?

      - get a mono camera; which mono camera? and is there any indication
      how much better results I could get?

      - I don't think longer subs are an option, 20 minutes is already pretty long

      - faster OTA? (like a hyperstar or boren-simon)

      - accept that my skies are terrible and accept that i will only get to
      image once a couple of months
    • Stan
      ... Not entirely. Your best bet is to image nebula with narrow band filters. They are great equalizers in that they remove nearly all sky pollution while
      Message 2 of 29 , Mar 14, 2013
      • 0 Attachment
        --- Orlando Andico <orly.andico@...> wrote:
        > ... an extremely light-polluted location
        >.. AP under these conditions is severely compromised...

        Not entirely. Your best bet is to image nebula with narrow band filters. They are great equalizers in that they remove nearly all sky pollution while preserving most of the object's signal.

        > ... for filtration, I'm using a stacked IDAS LPS-V4 and Red #25

        That's inefficient. Get a real H-a filter.

        > What, you say, OSC?! yes only 1/4 of the pixels get any light.
        >... rated at 50% QE in H-a...

        Meaning actual QE = 0.5*0.25 = 12.5%, which is abysmal for CCD though not unusual for photographic film. So it's not entirely hopeless but you have really stacked the deck against yourself. BTW, be sure to only extract (and not interpolate) the Red pixels for processing. Or better yet - sell that cam and get a real astro-cam.

        > ... signal is at about 2400 - 4300 ADU.
        > That is only a 2.5dB dynamic range.

        I didn't bother confirming your computation but check that you used electrons and not ADU because dynamic range must be calculated from real physical events (photon/electrons). Also for a complete profile it is necessary to subtract the sky signal but include the sky noise (as well as read noise) and should also include the shot noise of the signal to characterize total noise.

        Although such a measure may have some validity, that definition of dynamic range is unusual and is not especially useful. For the most part, dynamic range is simply a rating of the detector's ability to capture an intensity range within a single sub-exp. But because astro imaging routinely uses multiple subs and the impact of real dynamic range varies with sampling, it often isn't a very important measure.

        > it looks like even 10-minute subs are already more than enough,
        > because the signal is well above the pedestal. True or false?

        By pedestal, do you mean the sky background level? I question your assessment of "more than enough" and (as stated above) am unconvinced by your "dynamic range" analysis. I looked at that image and could barely even make out the nebula so I think you need to reassess that analysis.

        > How can I get it higher? - is this a camera issue, a narrower-filter
        > issue, or... my skies just well and truly suck?

        Sky glow can be almost completely removed with a true narrow band filter (3-5nm), so it's not the sky, at least for emission nebula. Your cobbled filer is sub-par and should be replaced. But the biggest obstacle is the OSC camera. Get rid of it and get a real astro-cam. If you used a proper filter and real astro-cam then you could expect a 6-20 fold increase in sensitivity compared to your currect setup (depending on how bad your sky really is).

        Stan
      • Orlando Andico
        Thanks Stan. I was expecting that a mono camera would be the recommendation. This guy s results though are different - http://www.helixgate.net/XMvsXCM2.html
        Message 3 of 29 , Mar 14, 2013
        • 0 Attachment
          Thanks Stan. I was expecting that a mono camera would be the
          recommendation.

          This guy's results though are different -
          http://www.helixgate.net/XMvsXCM2.html

          notice that he's comparing an ST-2000XM and ST-2000XCM, at the same
          exposure length, and except for the reduced resolution (which isn't even
          that bad) there doesn't seem to be a 5-fold or more difference that you
          suggest.

          I would agree that the H-a filter is a bad job, and a true narrower-band
          one is definitely something on my list.

          However the results above seem to show that an OSC camera isn't that much
          worse than a mono camera...



          On Fri, Mar 15, 2013 at 12:14 AM, Stan <stan_ccd@...> wrote:

          > **
          >
          >
          > --- Orlando Andico <orly.andico@...> wrote:
          > > ... an extremely light-polluted location
          > >.. AP under these conditions is severely compromised...
          >
          > Not entirely. Your best bet is to image nebula with narrow band filters.
          > They are great equalizers in that they remove nearly all sky pollution
          > while preserving most of the object's signal.
          >
          > > ... for filtration, I'm using a stacked IDAS LPS-V4 and Red #25
          >
          > That's inefficient. Get a real H-a filter.
          >
          >
          > > What, you say, OSC?! yes only 1/4 of the pixels get any light.
          > >... rated at 50% QE in H-a...
          >
          > Meaning actual QE = 0.5*0.25 = 12.5%, which is abysmal for CCD though not
          > unusual for photographic film. So it's not entirely hopeless but you have
          > really stacked the deck against yourself. BTW, be sure to only extract (and
          > not interpolate) the Red pixels for processing. Or better yet - sell that
          > cam and get a real astro-cam.
          >
          > > ... signal is at about 2400 - 4300 ADU.
          > > That is only a 2.5dB dynamic range.
          >
          > I didn't bother confirming your computation but check that you used
          > electrons and not ADU because dynamic range must be calculated from real
          > physical events (photon/electrons). Also for a complete profile it is
          > necessary to subtract the sky signal but include the sky noise (as well as
          > read noise) and should also include the shot noise of the signal to
          > characterize total noise.
          >
          > Although such a measure may have some validity, that definition of dynamic
          > range is unusual and is not especially useful. For the most part, dynamic
          > range is simply a rating of the detector's ability to capture an intensity
          > range within a single sub-exp. But because astro imaging routinely uses
          > multiple subs and the impact of real dynamic range varies with sampling, it
          > often isn't a very important measure.
          >
          >
          > > it looks like even 10-minute subs are already more than enough,
          > > because the signal is well above the pedestal. True or false?
          >
          > By pedestal, do you mean the sky background level? I question your
          > assessment of "more than enough" and (as stated above) am unconvinced by
          > your "dynamic range" analysis. I looked at that image and could barely even
          > make out the nebula so I think you need to reassess that analysis.
          >
          >
          > > How can I get it higher? - is this a camera issue, a narrower-filter
          > > issue, or... my skies just well and truly suck?
          >
          > Sky glow can be almost completely removed with a true narrow band filter
          > (3-5nm), so it's not the sky, at least for emission nebula. Your cobbled
          > filer is sub-par and should be replaced. But the biggest obstacle is the
          > OSC camera. Get rid of it and get a real astro-cam. If you used a proper
          > filter and real astro-cam then you could expect a 6-20 fold increase in
          > sensitivity compared to your currect setup (depending on how bad your sky
          > really is).
          >
          > Stan
          >
          >
          >



          --
          Orlando Andico
          +65.6436.1577 | +65.8139.0251


          [Non-text portions of this message have been removed]
        • Stan
          ... That was a typo, I meant 6-10 fold: a high quality H-a filter will pass 98% of H-a, whereas that stack probably passes 85% at best, thus the signal would
          Message 4 of 29 , Mar 14, 2013
          • 0 Attachment
            > ... used a proper filter and real astro-cam then you could
            > expect a 6-20 fold increase in sensitivity...

            That was a typo, I meant 6-10 fold:

            a high quality H-a filter will pass 98% of H-a, whereas that stack probably passes 85% at best, thus the signal would improve by about 1.15x.

            a 5nm filter passes 25% as much sky glow as a 20nm filter (a 3nm passes on 15% as much). Thus the noise from the sky is greatly diminished. S/N improvement for the dimmest H-a object would be about 2x for a 5nm compared to 20nm, assuming read noise is negligible (bad assumption!). A 2x improvement in S/N = 4x more signal. But that is tempered by read noise, so a more realistic boost might be about 2.5x.

            An identical camera with an unfiltered CCD (non-OSC) increases H-a QE by 4x.

            So the potential gain in sensitivity would be:

            1.15*2.5*4 = 11.5x

            meaning a 10 minute exp with proper filter/cam = 100 min exp with your current equip.

            Stan
          • Orlando Andico
            Stan, I don t understand the part about the true mono camera being 4X as sensitive. I understand that for a given aperture, the object gets projected onto the
            Message 5 of 29 , Mar 14, 2013
            • 0 Attachment
              Stan, I don't understand the part about the true mono camera being 4X as
              sensitive.

              I understand that for a given aperture, the object gets projected onto the
              CCD surface, and if you use an OSC you are blocking off 3/4 of that surface
              (hence the 1/4 QE) but on a per-pixel basis, you'd still be getting the
              same photon flux into each pixel, right?

              (e.g. if I get 1 e- / second into each pixel, with a mono camera I get 4 e-
              / second, with an OSC I only get 1 e- / second -- but each pixel would
              still end up with the same number of electrons, so only the spatial
              resolution would suffer).



              On Fri, Mar 15, 2013 at 1:41 AM, Stan <stan_ccd@...> wrote:

              > **
              >
              >
              > > ... used a proper filter and real astro-cam then you could
              > > expect a 6-20 fold increase in sensitivity...
              >
              > That was a typo, I meant 6-10 fold:
              >
              > a high quality H-a filter will pass 98% of H-a, whereas that stack
              > probably passes 85% at best, thus the signal would improve by about 1.15x.
              >
              > a 5nm filter passes 25% as much sky glow as a 20nm filter (a 3nm passes on
              > 15% as much). Thus the noise from the sky is greatly diminished. S/N
              > improvement for the dimmest H-a object would be about 2x for a 5nm compared
              > to 20nm, assuming read noise is negligible (bad assumption!). A 2x
              > improvement in S/N = 4x more signal. But that is tempered by read noise, so
              > a more realistic boost might be about 2.5x.
              >
              > An identical camera with an unfiltered CCD (non-OSC) increases H-a QE by
              > 4x.
              >
              > So the potential gain in sensitivity would be:
              >
              > 1.15*2.5*4 = 11.5x
              >
              > meaning a 10 minute exp with proper filter/cam = 100 min exp with your
              > current equip.
              >
              > Stan
              >
              >
              >



              --
              Orlando Andico
              +65.6436.1577 | +65.8139.0251


              [Non-text portions of this message have been removed]
            • Ron Wodaski
              Those images are stretched, which is somewhat misleading when comparing them. The whole point of a better exposure is that if you are going to stretch it, you
              Message 6 of 29 , Mar 14, 2013
              • 0 Attachment
                Those images are stretched, which is somewhat misleading when comparing them. The whole point of a better exposure is that if you are going to stretch it, you can stretch it _more_ because it's better data. Stretching them both the same doesn't reveal those differences as well, although I can see that the color image is significantly worse than the monochrome one - the loss of dynamic range (a result of greater noise) is the most obvious difference, but it's not easy to 'see' dynamic range. <g>

                So to my view, yes, it's a lot worse, but there are better ways to visualize that. For example, if each image were processed based on noise, the color one would have to be dimmer, and the mono one would be brighter (which the text says was the case; the comparison of DDP results hides this difference).

                Ron Wodaski



                On Mar 14, 2013, at 10:30 AM, Orlando Andico <orly.andico@...> wrote:

                > Thanks Stan. I was expecting that a mono camera would be the
                > recommendation.
                >
                > This guy's results though are different -
                > http://www.helixgate.net/XMvsXCM2.html
                >
                > notice that he's comparing an ST-2000XM and ST-2000XCM, at the same
                > exposure length, and except for the reduced resolution (which isn't even
                > that bad) there doesn't seem to be a 5-fold or more difference that you
                > suggest.
                >
                > I would agree that the H-a filter is a bad job, and a true narrower-band
                > one is definitely something on my list.
                >
                > However the results above seem to show that an OSC camera isn't that much
                > worse than a mono camera...
                >
                >
                >
                > On Fri, Mar 15, 2013 at 12:14 AM, Stan <stan_ccd@...> wrote:
                >
                >> **
                >>
                >>
                >> --- Orlando Andico <orly.andico@...> wrote:
                >>> ... an extremely light-polluted location
                >>> .. AP under these conditions is severely compromised...
                >>
                >> Not entirely. Your best bet is to image nebula with narrow band filters.
                >> They are great equalizers in that they remove nearly all sky pollution
                >> while preserving most of the object's signal.
                >>
                >>> ... for filtration, I'm using a stacked IDAS LPS-V4 and Red #25
                >>
                >> That's inefficient. Get a real H-a filter.
                >>
                >>
                >>> What, you say, OSC?! yes only 1/4 of the pixels get any light.
                >>> ... rated at 50% QE in H-a...
                >>
                >> Meaning actual QE = 0.5*0.25 = 12.5%, which is abysmal for CCD though not
                >> unusual for photographic film. So it's not entirely hopeless but you have
                >> really stacked the deck against yourself. BTW, be sure to only extract (and
                >> not interpolate) the Red pixels for processing. Or better yet - sell that
                >> cam and get a real astro-cam.
                >>
                >>> ... signal is at about 2400 - 4300 ADU.
                >>> That is only a 2.5dB dynamic range.
                >>
                >> I didn't bother confirming your computation but check that you used
                >> electrons and not ADU because dynamic range must be calculated from real
                >> physical events (photon/electrons). Also for a complete profile it is
                >> necessary to subtract the sky signal but include the sky noise (as well as
                >> read noise) and should also include the shot noise of the signal to
                >> characterize total noise.
                >>
                >> Although such a measure may have some validity, that definition of dynamic
                >> range is unusual and is not especially useful. For the most part, dynamic
                >> range is simply a rating of the detector's ability to capture an intensity
                >> range within a single sub-exp. But because astro imaging routinely uses
                >> multiple subs and the impact of real dynamic range varies with sampling, it
                >> often isn't a very important measure.
                >>
                >>
                >>> it looks like even 10-minute subs are already more than enough,
                >>> because the signal is well above the pedestal. True or false?
                >>
                >> By pedestal, do you mean the sky background level? I question your
                >> assessment of "more than enough" and (as stated above) am unconvinced by
                >> your "dynamic range" analysis. I looked at that image and could barely even
                >> make out the nebula so I think you need to reassess that analysis.
                >>
                >>
                >>> How can I get it higher? - is this a camera issue, a narrower-filter
                >>> issue, or... my skies just well and truly suck?
                >>
                >> Sky glow can be almost completely removed with a true narrow band filter
                >> (3-5nm), so it's not the sky, at least for emission nebula. Your cobbled
                >> filer is sub-par and should be replaced. But the biggest obstacle is the
                >> OSC camera. Get rid of it and get a real astro-cam. If you used a proper
                >> filter and real astro-cam then you could expect a 6-20 fold increase in
                >> sensitivity compared to your currect setup (depending on how bad your sky
                >> really is).
                >>
                >> Stan
                >>
                >>
                >>
                >
                >
                >
                > --
                > Orlando Andico
                > +65.6436.1577 | +65.8139.0251
                >
                >
                > [Non-text portions of this message have been removed]
                >
                >
                >
                > ------------------------------------
                >
                >
                >
                >
              • Orlando Andico
                Thanks Ron. You are right, it s hard to see dynamic range. :D I ll take your word for it. So the next question is.. which mono camera? is QE that
                Message 7 of 29 , Mar 14, 2013
                • 0 Attachment
                  Thanks Ron. You are right, it's hard to see dynamic range. :D I'll
                  take your word for it.

                  So the next question is.. which mono camera? is QE that important? based
                  on my last thread, it "seems" that the Sony 694 cameras are useful
                  contenders due to the high QE and low read noise. But the chip isn't that
                  large (12.5 x 10 mm).

                  For my specific use case, would the 694 camera be leaps and bounds better
                  than a low QE interline CCD like an ST-2000? (which only has 31% QE at the
                  H-a frequency, half of the 694.. it also has almost 3X the read noise)


                  On Fri, Mar 15, 2013 at 1:46 AM, Ron Wodaski <yahoo@...> wrote:

                  > **
                  >
                  >
                  > Those images are stretched, which is somewhat misleading when comparing
                  > them. The whole point of a better exposure is that if you are going to
                  > stretch it, you can stretch it _more_ because it's better data. Stretching
                  > them both the same doesn't reveal those differences as well, although I can
                  > see that the color image is significantly worse than the monochrome one -
                  > the loss of dynamic range (a result of greater noise) is the most obvious
                  > difference, but it's not easy to 'see' dynamic range. <g>
                  >
                  > So to my view, yes, it's a lot worse, but there are better ways to
                  > visualize that. For example, if each image were processed based on noise,
                  > the color one would have to be dimmer, and the mono one would be brighter
                  > (which the text says was the case; the comparison of DDP results hides this
                  > difference).
                  >
                  > Ron Wodaski
                  >
                  >
                  > On Mar 14, 2013, at 10:30 AM, Orlando Andico <orly.andico@...>
                  > wrote:
                  >
                  > > Thanks Stan. I was expecting that a mono camera would be the
                  > > recommendation.
                  > >
                  > > This guy's results though are different -
                  > > http://www.helixgate.net/XMvsXCM2.html
                  > >
                  > > notice that he's comparing an ST-2000XM and ST-2000XCM, at the same
                  > > exposure length, and except for the reduced resolution (which isn't even
                  > > that bad) there doesn't seem to be a 5-fold or more difference that you
                  > > suggest.
                  > >
                  > > I would agree that the H-a filter is a bad job, and a true narrower-band
                  > > one is definitely something on my list.
                  > >
                  > > However the results above seem to show that an OSC camera isn't that much
                  > > worse than a mono camera...
                  > >
                  > >
                  > >
                  > > On Fri, Mar 15, 2013 at 12:14 AM, Stan <stan_ccd@...> wrote:
                  > >
                  > >> **
                  >
                  > >>
                  > >>
                  > >> --- Orlando Andico <orly.andico@...> wrote:
                  > >>> ... an extremely light-polluted location
                  > >>> .. AP under these conditions is severely compromised...
                  > >>
                  > >> Not entirely. Your best bet is to image nebula with narrow band filters.
                  > >> They are great equalizers in that they remove nearly all sky pollution
                  > >> while preserving most of the object's signal.
                  > >>
                  > >>> ... for filtration, I'm using a stacked IDAS LPS-V4 and Red #25
                  > >>
                  > >> That's inefficient. Get a real H-a filter.
                  > >>
                  > >>
                  > >>> What, you say, OSC?! yes only 1/4 of the pixels get any light.
                  > >>> ... rated at 50% QE in H-a...
                  > >>
                  > >> Meaning actual QE = 0.5*0.25 = 12.5%, which is abysmal for CCD though
                  > not
                  > >> unusual for photographic film. So it's not entirely hopeless but you
                  > have
                  > >> really stacked the deck against yourself. BTW, be sure to only extract
                  > (and
                  > >> not interpolate) the Red pixels for processing. Or better yet - sell
                  > that
                  > >> cam and get a real astro-cam.
                  > >>
                  > >>> ... signal is at about 2400 - 4300 ADU.
                  > >>> That is only a 2.5dB dynamic range.
                  > >>
                  > >> I didn't bother confirming your computation but check that you used
                  > >> electrons and not ADU because dynamic range must be calculated from real
                  > >> physical events (photon/electrons). Also for a complete profile it is
                  > >> necessary to subtract the sky signal but include the sky noise (as well
                  > as
                  > >> read noise) and should also include the shot noise of the signal to
                  > >> characterize total noise.
                  > >>
                  > >> Although such a measure may have some validity, that definition of
                  > dynamic
                  > >> range is unusual and is not especially useful. For the most part,
                  > dynamic
                  > >> range is simply a rating of the detector's ability to capture an
                  > intensity
                  > >> range within a single sub-exp. But because astro imaging routinely uses
                  > >> multiple subs and the impact of real dynamic range varies with
                  > sampling, it
                  > >> often isn't a very important measure.
                  > >>
                  > >>
                  > >>> it looks like even 10-minute subs are already more than enough,
                  > >>> because the signal is well above the pedestal. True or false?
                  > >>
                  > >> By pedestal, do you mean the sky background level? I question your
                  > >> assessment of "more than enough" and (as stated above) am unconvinced by
                  > >> your "dynamic range" analysis. I looked at that image and could barely
                  > even
                  > >> make out the nebula so I think you need to reassess that analysis.
                  > >>
                  > >>
                  > >>> How can I get it higher? - is this a camera issue, a narrower-filter
                  > >>> issue, or... my skies just well and truly suck?
                  > >>
                  > >> Sky glow can be almost completely removed with a true narrow band filter
                  > >> (3-5nm), so it's not the sky, at least for emission nebula. Your cobbled
                  > >> filer is sub-par and should be replaced. But the biggest obstacle is the
                  > >> OSC camera. Get rid of it and get a real astro-cam. If you used a proper
                  > >> filter and real astro-cam then you could expect a 6-20 fold increase in
                  > >> sensitivity compared to your currect setup (depending on how bad your
                  > sky
                  > >> really is).
                  > >>
                  > >> Stan
                  > >>
                  > >>
                  > >>
                  > >
                  > >
                  > >
                  > > --
                  > > Orlando Andico
                  > > +65.6436.1577 | +65.8139.0251
                  > >
                  > >
                  > > [Non-text portions of this message have been removed]
                  > >
                  > >
                  > >
                  > > ------------------------------------
                  > >
                  > >
                  > >
                  > >
                  >
                  >
                  >



                  --
                  Orlando Andico
                  +65.6436.1577 | +65.8139.0251


                  [Non-text portions of this message have been removed]
                • Ron Wodaski
                  The short answer would be that a mono camera would have 4x the pixels, so would capture 4x the light. Both chips have the same _total area_, it s just that 3/4
                  Message 8 of 29 , Mar 14, 2013
                  • 0 Attachment
                    The short answer would be that a mono camera would have 4x the pixels, so would capture 4x the light. Both chips have the same _total area_, it's just that 3/4 of the area is unused with a color cam. Yes, the resolution of the final product is lower, but that's an arbitrary choice. You could display the result at any resolution you want to. The bottom line is that incoming signal gets wasted.

                    Ron Wodaski



                    On Mar 14, 2013, at 10:46 AM, Orlando Andico <orly.andico@...> wrote:

                    > Stan, I don't understand the part about the true mono camera being 4X as
                    > sensitive.
                    >
                    > I understand that for a given aperture, the object gets projected onto the
                    > CCD surface, and if you use an OSC you are blocking off 3/4 of that surface
                    > (hence the 1/4 QE) but on a per-pixel basis, you'd still be getting the
                    > same photon flux into each pixel, right?
                    >
                    > (e.g. if I get 1 e- / second into each pixel, with a mono camera I get 4 e-
                    > / second, with an OSC I only get 1 e- / second -- but each pixel would
                    > still end up with the same number of electrons, so only the spatial
                    > resolution would suffer).
                    >
                    >
                    >
                    > On Fri, Mar 15, 2013 at 1:41 AM, Stan <stan_ccd@...> wrote:
                    >
                    >> **
                    >>
                    >>
                    >>> ... used a proper filter and real astro-cam then you could
                    >>> expect a 6-20 fold increase in sensitivity...
                    >>
                    >> That was a typo, I meant 6-10 fold:
                    >>
                    >> a high quality H-a filter will pass 98% of H-a, whereas that stack
                    >> probably passes 85% at best, thus the signal would improve by about 1.15x.
                    >>
                    >> a 5nm filter passes 25% as much sky glow as a 20nm filter (a 3nm passes on
                    >> 15% as much). Thus the noise from the sky is greatly diminished. S/N
                    >> improvement for the dimmest H-a object would be about 2x for a 5nm compared
                    >> to 20nm, assuming read noise is negligible (bad assumption!). A 2x
                    >> improvement in S/N = 4x more signal. But that is tempered by read noise, so
                    >> a more realistic boost might be about 2.5x.
                    >>
                    >> An identical camera with an unfiltered CCD (non-OSC) increases H-a QE by
                    >> 4x.
                    >>
                    >> So the potential gain in sensitivity would be:
                    >>
                    >> 1.15*2.5*4 = 11.5x
                    >>
                    >> meaning a 10 minute exp with proper filter/cam = 100 min exp with your
                    >> current equip.
                    >>
                    >> Stan
                    >>
                    >>
                    >>
                    >
                    >
                    >
                    > --
                    > Orlando Andico
                    > +65.6436.1577 | +65.8139.0251
                    >
                    >
                    > [Non-text portions of this message have been removed]
                    >
                    >
                    >
                    > ------------------------------------
                    >
                    >
                    >
                    >
                  • Ron Wodaski
                    Well, it s only hard because it s unfamiliar. You can train yourself to see the differences that relate to dynamic range. Re: which mono camera? How much
                    Message 9 of 29 , Mar 14, 2013
                    • 0 Attachment
                      Well, it's only hard because it's unfamiliar. You can train yourself to see the differences that relate to dynamic range. <g>

                      Re: which mono camera?

                      How much do you want to spend? There are choices from cheap to $50k and up. You get what you pay for, more or less, with the usual deal: you don't get twice as much if you spend twice as much, there is a law of diminishing returns involved.

                      The key factors in camera performance are:

                      * QE (but it's not the be-all, end-all goal that it might seem to be)

                      * Low read noise (one part of the dynamic range equation)

                      * Large well depth (the other part of the dynamic range equation)

                      * Dark current noise (since you have to measure dark current, there is uncertainty (noise) in the measurement, and thus lower dark current is better)

                      At the high end, you might get a 95% QE back-illuminated chip with ultra-low read noise (< 1e-), a huge full-well capacity (800k and up), and extreme cooling (say to -90C and dark current way, way down at the 1e-/hour level). That will run you a pretty penny, but it will indeed be leaps and bounds better.

                      Short of such a pro-level camera, how many leaps and how many bounds better you can get depends nearly entirely on your budget. If you can set a budget range, then it's possible to offer some ideas. (Maybe not from me, since I now work mostly with pro-level equipment, but there is plenty of experience here at various price points.)

                      Whatever you buy, look at it from at least the above perspectives, then add on things like chip size, how much backfocus the camera design eats up, weight, etc. to get to something that works for you.

                      Ron Wodaski



                      On Mar 14, 2013, at 10:50 AM, Orlando Andico <orly.andico@...> wrote:

                      > Thanks Ron. You are right, it's hard to see dynamic range. :D I'll
                      > take your word for it.
                      >
                      > So the next question is.. which mono camera? is QE that important? based
                      > on my last thread, it "seems" that the Sony 694 cameras are useful
                      > contenders due to the high QE and low read noise. But the chip isn't that
                      > large (12.5 x 10 mm).
                      >
                      > For my specific use case, would the 694 camera be leaps and bounds better
                      > than a low QE interline CCD like an ST-2000? (which only has 31% QE at the
                      > H-a frequency, half of the 694.. it also has almost 3X the read noise)
                      >
                      >
                      > On Fri, Mar 15, 2013 at 1:46 AM, Ron Wodaski <yahoo@...> wrote:
                      >
                      >> **
                      >>
                      >>
                      >> Those images are stretched, which is somewhat misleading when comparing
                      >> them. The whole point of a better exposure is that if you are going to
                      >> stretch it, you can stretch it _more_ because it's better data. Stretching
                      >> them both the same doesn't reveal those differences as well, although I can
                      >> see that the color image is significantly worse than the monochrome one -
                      >> the loss of dynamic range (a result of greater noise) is the most obvious
                      >> difference, but it's not easy to 'see' dynamic range. <g>
                      >>
                      >> So to my view, yes, it's a lot worse, but there are better ways to
                      >> visualize that. For example, if each image were processed based on noise,
                      >> the color one would have to be dimmer, and the mono one would be brighter
                      >> (which the text says was the case; the comparison of DDP results hides this
                      >> difference).
                      >>
                      >> Ron Wodaski
                      >>
                      >>
                      >> On Mar 14, 2013, at 10:30 AM, Orlando Andico <orly.andico@...>
                      >> wrote:
                      >>
                      >>> Thanks Stan. I was expecting that a mono camera would be the
                      >>> recommendation.
                      >>>
                      >>> This guy's results though are different -
                      >>> http://www.helixgate.net/XMvsXCM2.html
                      >>>
                      >>> notice that he's comparing an ST-2000XM and ST-2000XCM, at the same
                      >>> exposure length, and except for the reduced resolution (which isn't even
                      >>> that bad) there doesn't seem to be a 5-fold or more difference that you
                      >>> suggest.
                      >>>
                      >>> I would agree that the H-a filter is a bad job, and a true narrower-band
                      >>> one is definitely something on my list.
                      >>>
                      >>> However the results above seem to show that an OSC camera isn't that much
                      >>> worse than a mono camera...
                      >>>
                      >>>
                      >>>
                      >>> On Fri, Mar 15, 2013 at 12:14 AM, Stan <stan_ccd@...> wrote:
                      >>>
                      >>>> **
                      >>
                      >>>>
                      >>>>
                      >>>> --- Orlando Andico <orly.andico@...> wrote:
                      >>>>> ... an extremely light-polluted location
                      >>>>> .. AP under these conditions is severely compromised...
                      >>>>
                      >>>> Not entirely. Your best bet is to image nebula with narrow band filters.
                      >>>> They are great equalizers in that they remove nearly all sky pollution
                      >>>> while preserving most of the object's signal.
                      >>>>
                      >>>>> ... for filtration, I'm using a stacked IDAS LPS-V4 and Red #25
                      >>>>
                      >>>> That's inefficient. Get a real H-a filter.
                      >>>>
                      >>>>
                      >>>>> What, you say, OSC?! yes only 1/4 of the pixels get any light.
                      >>>>> ... rated at 50% QE in H-a...
                      >>>>
                      >>>> Meaning actual QE = 0.5*0.25 = 12.5%, which is abysmal for CCD though
                      >> not
                      >>>> unusual for photographic film. So it's not entirely hopeless but you
                      >> have
                      >>>> really stacked the deck against yourself. BTW, be sure to only extract
                      >> (and
                      >>>> not interpolate) the Red pixels for processing. Or better yet - sell
                      >> that
                      >>>> cam and get a real astro-cam.
                      >>>>
                      >>>>> ... signal is at about 2400 - 4300 ADU.
                      >>>>> That is only a 2.5dB dynamic range.
                      >>>>
                      >>>> I didn't bother confirming your computation but check that you used
                      >>>> electrons and not ADU because dynamic range must be calculated from real
                      >>>> physical events (photon/electrons). Also for a complete profile it is
                      >>>> necessary to subtract the sky signal but include the sky noise (as well
                      >> as
                      >>>> read noise) and should also include the shot noise of the signal to
                      >>>> characterize total noise.
                      >>>>
                      >>>> Although such a measure may have some validity, that definition of
                      >> dynamic
                      >>>> range is unusual and is not especially useful. For the most part,
                      >> dynamic
                      >>>> range is simply a rating of the detector's ability to capture an
                      >> intensity
                      >>>> range within a single sub-exp. But because astro imaging routinely uses
                      >>>> multiple subs and the impact of real dynamic range varies with
                      >> sampling, it
                      >>>> often isn't a very important measure.
                      >>>>
                      >>>>
                      >>>>> it looks like even 10-minute subs are already more than enough,
                      >>>>> because the signal is well above the pedestal. True or false?
                      >>>>
                      >>>> By pedestal, do you mean the sky background level? I question your
                      >>>> assessment of "more than enough" and (as stated above) am unconvinced by
                      >>>> your "dynamic range" analysis. I looked at that image and could barely
                      >> even
                      >>>> make out the nebula so I think you need to reassess that analysis.
                      >>>>
                      >>>>
                      >>>>> How can I get it higher? - is this a camera issue, a narrower-filter
                      >>>>> issue, or... my skies just well and truly suck?
                      >>>>
                      >>>> Sky glow can be almost completely removed with a true narrow band filter
                      >>>> (3-5nm), so it's not the sky, at least for emission nebula. Your cobbled
                      >>>> filer is sub-par and should be replaced. But the biggest obstacle is the
                      >>>> OSC camera. Get rid of it and get a real astro-cam. If you used a proper
                      >>>> filter and real astro-cam then you could expect a 6-20 fold increase in
                      >>>> sensitivity compared to your currect setup (depending on how bad your
                      >> sky
                      >>>> really is).
                      >>>>
                      >>>> Stan
                      >>>>
                      >>>>
                      >>>>
                      >>>
                      >>>
                      >>>
                      >>> --
                      >>> Orlando Andico
                      >>> +65.6436.1577 | +65.8139.0251
                      >>>
                      >>>
                      >>> [Non-text portions of this message have been removed]
                      >>>
                      >>>
                      >>>
                      >>> ------------------------------------
                      >>>
                      >>>
                      >>>
                      >>>
                      >>
                      >>
                      >>
                      >
                      >
                      >
                      > --
                      > Orlando Andico
                      > +65.6436.1577 | +65.8139.0251
                      >
                      >
                      > [Non-text portions of this message have been removed]
                      >
                      >
                      >
                      > ------------------------------------
                      >
                      >
                      >
                      >
                    • Stan
                      ... Not really. His setups are equivalent (same filter efficiency and pass band) so that the Bayer matrix is the primary variable. A 4x difference in
                      Message 10 of 29 , Mar 14, 2013
                      • 0 Attachment
                        --- Orlando Andico <orly.andico@...> wrote:
                        > This guy's results though are different -
                        > http://www.helixgate.net/XMvsXCM2.html

                        Not really. His setups are equivalent (same filter efficiency and pass band) so that the Bayer matrix is the primary variable. A 4x difference in "sensitivity" results in a 2x difference in S/N and a careful look at those images exhibits just about that much – there is definitely better limiting mag and contrast detection in the non-OSC.

                        However, those exps may a bit short for narrow band (depending on camera noise and f-ratio) so read noise may have affected his results and understated the potential difference. But to my eye there is a significant difference anyway. BTW, a more realistic compare, that he neglected, would be to take a 4x longer exp with the OSC to see if it is any better than the mono (I predict not).

                        Stan
                      • Orlando Andico
                        Thanks for the reply, Ron. Sadly, I am not made of money, so those fancy back-illuminated chips are out. This is after all a hobby. So.. if I m looking at a
                        Message 11 of 29 , Mar 14, 2013
                        • 0 Attachment
                          Thanks for the reply, Ron.

                          Sadly, I am not made of money, so those fancy back-illuminated chips are
                          out. This is after all a hobby.

                          So.. if I'm looking at a $2K to $3K camera, which is really the entry-level
                          for mono, that's where I'm worried that anything I could get, would not be
                          leaps-and-bounds better...

                          I'm not concerned about things like back-focus (since I'm not going to use
                          an OAG, and my OTA has a lot of back-focus, far more than enough for things
                          like a filter wheel)

                          I am also not averse to used.. the ST-10XME seemed like an interesting
                          choice (KAF-3200) - a new KAF-3200 camera is over my budget.



                          On Fri, Mar 15, 2013 at 2:10 AM, Ron Wodaski <yahoo@...> wrote:

                          > **
                          >
                          >
                          > Well, it's only hard because it's unfamiliar. You can train yourself to
                          > see the differences that relate to dynamic range. <g>
                          >
                          > Re: which mono camera?
                          >
                          > How much do you want to spend? There are choices from cheap to $50k and
                          > up. You get what you pay for, more or less, with the usual deal: you don't
                          > get twice as much if you spend twice as much, there is a law of diminishing
                          > returns involved.
                          >
                          > The key factors in camera performance are:
                          >
                          > * QE (but it's not the be-all, end-all goal that it might seem to be)
                          >
                          > * Low read noise (one part of the dynamic range equation)
                          >
                          > * Large well depth (the other part of the dynamic range equation)
                          >
                          > * Dark current noise (since you have to measure dark current, there is
                          > uncertainty (noise) in the measurement, and thus lower dark current is
                          > better)
                          >
                          > At the high end, you might get a 95% QE back-illuminated chip with
                          > ultra-low read noise (< 1e-), a huge full-well capacity (800k and up), and
                          > extreme cooling (say to -90C and dark current way, way down at the 1e-/hour
                          > level). That will run you a pretty penny, but it will indeed be leaps and
                          > bounds better.
                          >
                          > Short of such a pro-level camera, how many leaps and how many bounds
                          > better you can get depends nearly entirely on your budget. If you can set a
                          > budget range, then it's possible to offer some ideas. (Maybe not from me,
                          > since I now work mostly with pro-level equipment, but there is plenty of
                          > experience here at various price points.)
                          >
                          > Whatever you buy, look at it from at least the above perspectives, then
                          > add on things like chip size, how much backfocus the camera design eats up,
                          > weight, etc. to get to something that works for you.
                          >
                          > Ron Wodaski
                          >
                          >
                          > On Mar 14, 2013, at 10:50 AM, Orlando Andico <orly.andico@...>
                          > wrote:
                          >
                          > > Thanks Ron. You are right, it's hard to see dynamic range. :D I'll
                          > > take your word for it.
                          > >
                          > > So the next question is.. which mono camera? is QE that important? based
                          > > on my last thread, it "seems" that the Sony 694 cameras are useful
                          > > contenders due to the high QE and low read noise. But the chip isn't that
                          > > large (12.5 x 10 mm).
                          > >
                          > > For my specific use case, would the 694 camera be leaps and bounds better
                          > > than a low QE interline CCD like an ST-2000? (which only has 31% QE at
                          > the
                          > > H-a frequency, half of the 694.. it also has almost 3X the read noise)
                          > >
                          > >
                          > > On Fri, Mar 15, 2013 at 1:46 AM, Ron Wodaski <yahoo@...> wrote:
                          > >
                          > >> **
                          >
                          > >>
                          > >>
                          > >> Those images are stretched, which is somewhat misleading when comparing
                          > >> them. The whole point of a better exposure is that if you are going to
                          > >> stretch it, you can stretch it _more_ because it's better data.
                          > Stretching
                          > >> them both the same doesn't reveal those differences as well, although I
                          > can
                          > >> see that the color image is significantly worse than the monochrome one
                          > -
                          > >> the loss of dynamic range (a result of greater noise) is the most
                          > obvious
                          > >> difference, but it's not easy to 'see' dynamic range. <g>
                          > >>
                          > >> So to my view, yes, it's a lot worse, but there are better ways to
                          > >> visualize that. For example, if each image were processed based on
                          > noise,
                          > >> the color one would have to be dimmer, and the mono one would be
                          > brighter
                          > >> (which the text says was the case; the comparison of DDP results hides
                          > this
                          > >> difference).
                          > >>
                          > >> Ron Wodaski
                          > >>
                          > >>
                          > >> On Mar 14, 2013, at 10:30 AM, Orlando Andico <orly.andico@...>
                          > >> wrote:
                          > >>
                          > >>> Thanks Stan. I was expecting that a mono camera would be the
                          > >>> recommendation.
                          > >>>
                          > >>> This guy's results though are different -
                          > >>> http://www.helixgate.net/XMvsXCM2.html
                          > >>>
                          > >>> notice that he's comparing an ST-2000XM and ST-2000XCM, at the same
                          > >>> exposure length, and except for the reduced resolution (which isn't
                          > even
                          > >>> that bad) there doesn't seem to be a 5-fold or more difference that you
                          > >>> suggest.
                          > >>>
                          > >>> I would agree that the H-a filter is a bad job, and a true
                          > narrower-band
                          > >>> one is definitely something on my list.
                          > >>>
                          > >>> However the results above seem to show that an OSC camera isn't that
                          > much
                          > >>> worse than a mono camera...
                          > >>>
                          > >>>
                          > >>>
                          > >>> On Fri, Mar 15, 2013 at 12:14 AM, Stan <stan_ccd@...> wrote:
                          > >>>
                          > >>>> **
                          > >>
                          > >>>>
                          > >>>>
                          > >>>> --- Orlando Andico <orly.andico@...> wrote:
                          > >>>>> ... an extremely light-polluted location
                          > >>>>> .. AP under these conditions is severely compromised...
                          > >>>>
                          > >>>> Not entirely. Your best bet is to image nebula with narrow band
                          > filters.
                          > >>>> They are great equalizers in that they remove nearly all sky pollution
                          > >>>> while preserving most of the object's signal.
                          > >>>>
                          > >>>>> ... for filtration, I'm using a stacked IDAS LPS-V4 and Red #25
                          > >>>>
                          > >>>> That's inefficient. Get a real H-a filter.
                          > >>>>
                          > >>>>
                          > >>>>> What, you say, OSC?! yes only 1/4 of the pixels get any light.
                          > >>>>> ... rated at 50% QE in H-a...
                          > >>>>
                          > >>>> Meaning actual QE = 0.5*0.25 = 12.5%, which is abysmal for CCD though
                          > >> not
                          > >>>> unusual for photographic film. So it's not entirely hopeless but you
                          > >> have
                          > >>>> really stacked the deck against yourself. BTW, be sure to only extract
                          > >> (and
                          > >>>> not interpolate) the Red pixels for processing. Or better yet - sell
                          > >> that
                          > >>>> cam and get a real astro-cam.
                          > >>>>
                          > >>>>> ... signal is at about 2400 - 4300 ADU.
                          > >>>>> That is only a 2.5dB dynamic range.
                          > >>>>
                          > >>>> I didn't bother confirming your computation but check that you used
                          > >>>> electrons and not ADU because dynamic range must be calculated from
                          > real
                          > >>>> physical events (photon/electrons). Also for a complete profile it is
                          > >>>> necessary to subtract the sky signal but include the sky noise (as
                          > well
                          > >> as
                          > >>>> read noise) and should also include the shot noise of the signal to
                          > >>>> characterize total noise.
                          > >>>>
                          > >>>> Although such a measure may have some validity, that definition of
                          > >> dynamic
                          > >>>> range is unusual and is not especially useful. For the most part,
                          > >> dynamic
                          > >>>> range is simply a rating of the detector's ability to capture an
                          > >> intensity
                          > >>>> range within a single sub-exp. But because astro imaging routinely
                          > uses
                          > >>>> multiple subs and the impact of real dynamic range varies with
                          > >> sampling, it
                          > >>>> often isn't a very important measure.
                          > >>>>
                          > >>>>
                          > >>>>> it looks like even 10-minute subs are already more than enough,
                          > >>>>> because the signal is well above the pedestal. True or false?
                          > >>>>
                          > >>>> By pedestal, do you mean the sky background level? I question your
                          > >>>> assessment of "more than enough" and (as stated above) am unconvinced
                          > by
                          > >>>> your "dynamic range" analysis. I looked at that image and could barely
                          > >> even
                          > >>>> make out the nebula so I think you need to reassess that analysis.
                          > >>>>
                          > >>>>
                          > >>>>> How can I get it higher? - is this a camera issue, a narrower-filter
                          > >>>>> issue, or... my skies just well and truly suck?
                          > >>>>
                          > >>>> Sky glow can be almost completely removed with a true narrow band
                          > filter
                          > >>>> (3-5nm), so it's not the sky, at least for emission nebula. Your
                          > cobbled
                          > >>>> filer is sub-par and should be replaced. But the biggest obstacle is
                          > the
                          > >>>> OSC camera. Get rid of it and get a real astro-cam. If you used a
                          > proper
                          > >>>> filter and real astro-cam then you could expect a 6-20 fold increase
                          > in
                          > >>>> sensitivity compared to your currect setup (depending on how bad your
                          > >> sky
                          > >>>> really is).
                          > >>>>
                          > >>>> Stan
                          > >>>>
                          > >>>>
                          > >>>>
                          > >>>
                          > >>>
                          > >>>
                          > >>> --
                          > >>> Orlando Andico
                          > >>> +65.6436.1577 | +65.8139.0251
                          > >>>
                          > >>>
                          > >>> [Non-text portions of this message have been removed]
                          > >>>
                          > >>>
                          > >>>
                          > >>> ------------------------------------
                          > >>>
                          > >>>
                          > >>>
                          > >>>
                          > >>
                          > >>
                          > >>
                          > >
                          > >
                          > >
                          > > --
                          > > Orlando Andico
                          > > +65.6436.1577 | +65.8139.0251
                          > >
                          > >
                          > > [Non-text portions of this message have been removed]
                          > >
                          > >
                          > >
                          > > ------------------------------------
                          > >
                          > >
                          > >
                          > >
                          >
                          >
                          >



                          --
                          Orlando Andico
                          +65.6436.1577 | +65.8139.0251


                          [Non-text portions of this message have been removed]
                        • Stan
                          ... The true mission and metric of a camera is how much information it can capture from the virtual image presented to it. Photons = information and basic
                          Message 12 of 29 , Mar 14, 2013
                          • 0 Attachment
                            --- Orlando Andico <orly.andico@...> wrote:
                            > I don't understand the part about the true mono camera
                            > being 4X as sensitive.

                            The true mission and metric of a camera is how much information it can capture from the virtual image presented to it. Photons = information and basic photon accounting reveals that failing to detect 3 out of every 4 photons from an object produces a quantum (photon) efficiency = 0.25x.

                            Stan
                          • Stan
                            ... Actually I meant 4x more sub-exps to produce 4x longer total exp. Taking longer sub-exps for only one camera would unbalance the affects of read noise and
                            Message 13 of 29 , Mar 14, 2013
                            • 0 Attachment
                              > a more realistic compare, that he neglected, would be to take
                              > a 4x longer exp with the OSC to see if it is any better...

                              Actually I meant 4x more sub-exps to produce 4x longer total exp.

                              Taking longer sub-exps for only one camera would unbalance the affects of read noise and favor the cam with the longer exps.

                              Stan
                            • Orlando Andico
                              There is another option that obviously appeals to the penny-pincher (and DIY er) in me. An equipment liquidator on ebay is selling these old Quantix microscopy
                              Message 14 of 29 , Mar 14, 2013
                              • 0 Attachment
                                There is another option that obviously appeals to the penny-pincher (and
                                DIY'er) in me.

                                An equipment liquidator on ebay is selling these old Quantix microscopy
                                cameras for really low prices.

                                They are using KAF-6303 chips, which have 9um pixels, a pretty decent full
                                well (85000 e-) compared to the Sony or 8300 chips, The QE is about 65%.
                                The dark current is pretty high (0.3 e-/sec @ 0) and the read noise is 11
                                e-.

                                I'm assuming the RMS dark noise is the square root, and total noise is sqrt
                                (dark^2 + read^2). For this particular camera, total noise would be about
                                17e- at 600 seconds. So full well / noise (dynamic range?) = 37 dB.

                                If we look at a typical 8300, the QE is 56%, the well depth much less
                                (25000 e-), the dark current is less (0,02 e- @ - 15, or about 0,1 e-/sec @
                                0) and the read noise is 9 e-.. At 600 seconds, total would be about 12e-
                                of noise, and full well / noise = 33 dB.

                                Or if we look at a 694. QE 77%, well depth 20000 e-, dark current ~ 0.004
                                e- @ 0, read noise 5 e-. At 600 seconds total would be ~ 6 e- and DR would
                                be about 35 dB.

                                Or are all of these cameras broadly comparable?


                                [Non-text portions of this message have been removed]
                              • Stan
                                ... This can be a great deal (I obtained a used microscopy EMCCD from a similar liquadator for 10% of retail). I didn;t look at that one, but most of lab CCDs
                                Message 15 of 29 , Mar 14, 2013
                                • 0 Attachment
                                  --- Orlando Andico <orly.andico@...> wrote:
                                  > There is another option ...
                                  > An equipment liquidator on ebay is selling these old
                                  > Quantix microscopy cameras for really low prices...

                                  This can be a great deal (I obtained a used microscopy EMCCD from a similar liquadator for 10% of retail).

                                  I didn;t look at that one, but most of lab CCDs use exotic interfaces such as "Camera Link" (expensive and awful interface) or a dedicated PCI board that requires a computer that is hardware compatible. And often the driver is 32 bit only. This eliminates most laptops and many modern desktops. So in addition to one of those cameas you may also need a comptible desktop computer (also can be bought used).

                                  Additionally, many of these cameras require expensive software that is ill-suited for astro. I had to write my own camera control software, which took ahwile but was worth it.

                                  Stan
                                • Orlando Andico
                                  Hi Stan, These microscopy cameras use a PCI interface, and they come with their own capture software BUT can also be driven by MaximDL Pro. A desktop or laptop
                                  Message 16 of 29 , Mar 14, 2013
                                  • 0 Attachment
                                    Hi Stan,

                                    These microscopy cameras use a PCI interface, and they come with their own
                                    capture software BUT can also be driven by MaximDL Pro. A desktop or laptop
                                    with big dock are needed, and you're right I believe they are 32-bit only.

                                    I actually got an Apogee KX camera which was being sold as a KX32ME
                                    (KAF-3200ME chip) and I got it working with MaximDL -- but it turned out to
                                    be a KX85 (Sony ICX085 chip - much less desirable) and I had to send it
                                    back.

                                    But that aside, given the cost constraints and my previous calculations,
                                    which of the cameras would be recommended? or are they all more-or-less
                                    equivalent?


                                    On Fri, Mar 15, 2013 at 3:45 AM, Stan <stan_ccd@...> wrote:

                                    > **
                                    >
                                    >
                                    > --- Orlando Andico <orly.andico@...> wrote:
                                    > > There is another option ...
                                    > > An equipment liquidator on ebay is selling these old
                                    > > Quantix microscopy cameras for really low prices...
                                    >
                                    > This can be a great deal (I obtained a used microscopy EMCCD from a
                                    > similar liquadator for 10% of retail).
                                    >
                                    > I didn;t look at that one, but most of lab CCDs use exotic interfaces such
                                    > as "Camera Link" (expensive and awful interface) or a dedicated PCI board
                                    > that requires a computer that is hardware compatible. And often the driver
                                    > is 32 bit only. This eliminates most laptops and many modern desktops. So
                                    > in addition to one of those cameas you may also need a comptible desktop
                                    > computer (also can be bought used).
                                    >
                                    > Additionally, many of these cameras require expensive software that is
                                    > ill-suited for astro. I had to write my own camera control software, which
                                    > took ahwile but was worth it.
                                    >
                                    > Stan
                                    >
                                    >
                                    >



                                    --
                                    Orlando Andico
                                    +65.6436.1577 | +65.8139.0251


                                    [Non-text portions of this message have been removed]
                                  • Orly
                                    I just measured my NELM with Samir Kharusi s formula Mag/sq arc-sec = 13.93+2.5*log10(seconds to mid histogram at ISO 800 and f4) and came up with 15 seconds,
                                    Message 17 of 29 , Mar 15, 2013
                                    • 0 Attachment
                                      I just measured my NELM with Samir Kharusi's formula

                                      Mag/sq arc-sec = 13.93+2.5*log10(seconds to mid histogram at ISO 800 and f4)


                                      and came up with 15 seconds, which is 16.87 mag / sq arc-sec.

                                      from this formula
                                      nelm = 7.93 - 5 log(10^(4.316-(bmpsas/5)) + 1)

                                      i figured my NELM is... 3.

                                      given such a (terrible) amount of sky fog, is my goal still practical? with a mono camera and 3- or 5-nm Ha filter, what sort of exposure lengths should i expect?



                                      --- In ccd-newastro@yahoogroups.com, "Stan" <stan_ccd@...> wrote:
                                      >
                                      > > a more realistic compare, that he neglected, would be to take
                                      > > a 4x longer exp with the OSC to see if it is any better...
                                      >
                                      > Actually I meant 4x more sub-exps to produce 4x longer total exp.
                                      >
                                      > Taking longer sub-exps for only one camera would unbalance the affects of read noise and favor the cam with the longer exps.
                                      >
                                      > Stan
                                      >
                                    • Stan
                                      ... Essentially all cameras based on any particular CCD are equivalent in terms of QE, read noise and dark current (per temperature) as there is little that a
                                      Message 18 of 29 , Mar 15, 2013
                                      • 0 Attachment
                                        --- Orlando Andico <orly.andico@...> wrote:
                                        > ... or are they all more-or-less equivalent?

                                        Essentially all cameras based on any particular CCD are equivalent in terms of QE, read noise and dark current (per temperature) as there is little that a camera manufacturer can do about those attributes other than screw them up. The things that differentiates such cameras are form factors (e.g. lab cameras are usually large and heavy), built-in features (many lab cameras lack an internal shutter, which can be very problematic for flat fields), cooling, lens mount (usually c-mount or proprietary for lab cameras), data transmission interface (and speed of download), software drivers and compatibilities. There are also tangential but potentially important issues such as repair costs (most lab cameras cost more to repair than a new amateur camera) and obsolescence (e.g. 32 or 32/64 PCI slots have become very rare in motherboards and that card will not work in a 64 PCI bit slot).

                                        If you are OK using an old desktop machine for the camera and can figure out some sort of robust shutter for flat fields then it might be worth pursuing. The included (or more likely downloadable) software should be sufficient for manual exposures (MaxIm could be used for semi-automation, though if you don't already have it, it is a significant added expense). But it is not for the weak... <g>

                                        Comparing different CCDs is more complicated because pixel size and noise characteristics have differing effects for particular scopes, filters, applications and techniques.

                                        Stan
                                      • Orly
                                        well... there is an active group of users on CN who have successfully added shutters to these Quantix cameras and are using them for AP. The cost of the
                                        Message 19 of 29 , Mar 15, 2013
                                        • 0 Attachment
                                          well... there is an active group of users on CN who have successfully added shutters to these Quantix cameras and are using them for AP.

                                          The cost of the Quantix 6303 + MaximDL Pro is only sufficient to buy an Atik 314L+ (ICX285 chip). So......



                                          --- In ccd-newastro@yahoogroups.com, "Stan" <stan_ccd@...> wrote:
                                          >
                                          > --- Orlando Andico <orly.andico@> wrote:
                                          > > ... or are they all more-or-less equivalent?
                                          >
                                          > Essentially all cameras based on any particular CCD are equivalent in terms of QE, read noise and dark current (per temperature) as there is little that a camera manufacturer can do about those attributes other than screw them up. The things that differentiates such cameras are form factors (e.g. lab cameras are usually large and heavy), built-in features (many lab cameras lack an internal shutter, which can be very problematic for flat fields), cooling, lens mount (usually c-mount or proprietary for lab cameras), data transmission interface (and speed of download), software drivers and compatibilities. There are also tangential but potentially important issues such as repair costs (most lab cameras cost more to repair than a new amateur camera) and obsolescence (e.g. 32 or 32/64 PCI slots have become very rare in motherboards and that card will not work in a 64 PCI bit slot).
                                          >
                                          > If you are OK using an old desktop machine for the camera and can figure out some sort of robust shutter for flat fields then it might be worth pursuing. The included (or more likely downloadable) software should be sufficient for manual exposures (MaxIm could be used for semi-automation, though if you don't already have it, it is a significant added expense). But it is not for the weak... <g>
                                          >
                                          > Comparing different CCDs is more complicated because pixel size and noise characteristics have differing effects for particular scopes, filters, applications and techniques.
                                          >
                                          > Stan
                                          >
                                        • Stan
                                          ... Do you have a link to that group? Stan
                                          Message 20 of 29 , Mar 15, 2013
                                          • 0 Attachment
                                            --- "Orly" <orly.andico@...> wrote:
                                            > there is an active group of users on CN...

                                            Do you have a link to that group?

                                            Stan
                                          • Orlando Andico
                                            VERY long thread.. http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/4619543/page/1/view/collapsed/sb/5/o/all/fpart/1 Most people end up using
                                            Message 21 of 29 , Mar 15, 2013
                                            • 0 Attachment
                                              VERY long thread..

                                              http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/4619543/page/1/view/collapsed/sb/5/o/all/fpart/1

                                              Most people end up using VS35 (Uniblitz) or Melles-Griot shutters. The
                                              camera already has the wires to trigger a shutter.



                                              On Sat, Mar 16, 2013 at 12:48 AM, Stan <stan_ccd@...> wrote:

                                              > **
                                              >
                                              >
                                              > --- "Orly" <orly.andico@...> wrote:
                                              > > there is an active group of users on CN...
                                              >
                                              > Do you have a link to that group?
                                              >
                                              > Stan
                                              >
                                              >
                                              >



                                              --
                                              Orlando Andico
                                              +65.6436.1577 | +65.8139.0251


                                              [Non-text portions of this message have been removed]
                                            • Ron Wodaski
                                              You can roll your own shutter with parts from Uniblitz (you need a shutter, and a driver to trigger it). Not exactly cheap, but workable. This assumes the
                                              Message 22 of 29 , Mar 15, 2013
                                              • 0 Attachment
                                                You can roll your own shutter with parts from Uniblitz (you need a shutter, and a driver to trigger it). Not exactly cheap, but workable. This assumes the camera has a relay connection to trigger an external shutter; many do.

                                                Ron Wodaski



                                                On Mar 15, 2013, at 9:09 AM, Stan <stan_ccd@...> wrote:

                                                > --- Orlando Andico <orly.andico@...> wrote:
                                                >> ... or are they all more-or-less equivalent?
                                                >
                                                > Essentially all cameras based on any particular CCD are equivalent in terms of QE, read noise and dark current (per temperature) as there is little that a camera manufacturer can do about those attributes other than screw them up. The things that differentiates such cameras are form factors (e.g. lab cameras are usually large and heavy), built-in features (many lab cameras lack an internal shutter, which can be very problematic for flat fields), cooling, lens mount (usually c-mount or proprietary for lab cameras), data transmission interface (and speed of download), software drivers and compatibilities. There are also tangential but potentially important issues such as repair costs (most lab cameras cost more to repair than a new amateur camera) and obsolescence (e.g. 32 or 32/64 PCI slots have become very rare in motherboards and that card will not work in a 64 PCI bit slot).
                                                >
                                                > If you are OK using an old desktop machine for the camera and can figure out some sort of robust shutter for flat fields then it might be worth pursuing. The included (or more likely downloadable) software should be sufficient for manual exposures (MaxIm could be used for semi-automation, though if you don't already have it, it is a significant added expense). But it is not for the weak... <g>
                                                >
                                                > Comparing different CCDs is more complicated because pixel size and noise characteristics have differing effects for particular scopes, filters, applications and techniques.
                                                >
                                                > Stan
                                                >
                                                >
                                                >
                                                >
                                                > ------------------------------------
                                                >
                                                >
                                                >
                                                >
                                              • Orlando Andico
                                                Ron, this specific camera already has a shutter driver. 24V trigger, 4V hold. So given that a new 6303 camera starts at about $7K, what the liquidator is
                                                Message 23 of 29 , Mar 15, 2013
                                                • 0 Attachment
                                                  Ron, this specific camera already has a shutter driver. 24V trigger, 4V
                                                  hold.

                                                  So given that a new 6303 camera starts at about $7K, what the liquidator is
                                                  asking ($1K) is reasonable? even when factoring in the machining work to
                                                  attach a shutter and the cost of MaximDL (although MaximDL is not
                                                  required.. it comes with software that can capture, then just use PHD to
                                                  guide).



                                                  On Sat, Mar 16, 2013 at 1:00 AM, Ron Wodaski <yahoo@...> wrote:

                                                  > **
                                                  >
                                                  >
                                                  > You can roll your own shutter with parts from Uniblitz (you need a
                                                  > shutter, and a driver to trigger it). Not exactly cheap, but workable. This
                                                  > assumes the camera has a relay connection to trigger an external shutter;
                                                  > many do.
                                                  >
                                                  > Ron Wodaski
                                                  >


                                                  [Non-text portions of this message have been removed]
                                                • Neil Fleming
                                                  IIRC, my NELM from Boston was 16.95.  I ve since moved my gear, but this shot was taken from the white light polluted area in Boston:
                                                  Message 24 of 29 , Mar 15, 2013
                                                  • 0 Attachment
                                                    IIRC, my NELM from Boston was 16.95.  I've since moved my gear, but this shot was taken from the "white" light polluted area in Boston:
                                                    http://www.flemingastrophotography.com/ic1396.html


                                                       ...Neil
                                                     
                                                    www.flemingastrophotography.com 
                                                    Direct from Boston - brilliant diamonds in pea soup
                                                    Also check out the astro_narrowbandYahoo group!



                                                    >________________________________
                                                    > From: Orly <orly.andico@...>
                                                    >To: ccd-newastro@yahoogroups.com
                                                    >Sent: Friday, March 15, 2013 12:04 PM
                                                    >Subject: [ccd-newastro] Re: looking for camera (or workflow) recommendation
                                                    >
                                                    >I just measured my NELM with Samir Kharusi's formula
                                                    >
                                                    >Mag/sq arc-sec = 13.93+2.5*log10(seconds to mid histogram at ISO 800 and f4)
                                                    >
                                                    >
                                                    >and came up with 15 seconds, which is 16.87 mag / sq arc-sec.
                                                    >
                                                    >from this formula
                                                    >nelm = 7.93 - 5 log(10^(4.316-(bmpsas/5)) + 1)
                                                    >
                                                    >i figured my NELM is... 3.
                                                    >
                                                    >given such a (terrible) amount of sky fog, is my goal still practical? with a mono camera and 3- or 5-nm Ha filter, what sort of exposure lengths should i expect?
                                                    >
                                                    >
                                                    >
                                                    >--- In ccd-newastro@yahoogroups.com, "Stan" <stan_ccd@...> wrote:
                                                    >>
                                                    >> > a more realistic compare, that he neglected, would be to take
                                                    >> > a 4x longer exp with the OSC to see if it is any better...
                                                    >>
                                                    >> Actually I meant 4x more sub-exps to produce 4x longer total exp.
                                                    >>
                                                    >> Taking longer sub-exps for only one camera would unbalance the affects of read noise and favor the cam with the longer exps.
                                                    >>
                                                    >> Stan
                                                    >>
                                                    >
                                                    >
                                                    >
                                                    >------------------------------------
                                                    >
                                                    >
                                                    >
                                                    >
                                                    >
                                                    >
                                                    >

                                                    [Non-text portions of this message have been removed]
                                                  • Ron Wodaski
                                                    Well, it s really up to you to determine both the value and whether you want to do the work. I can t help with that. As to value, I have not examined this
                                                    Message 25 of 29 , Mar 15, 2013
                                                    • 0 Attachment
                                                      Well, it's really up to you to determine both the value and whether you want to do the work. I can't help with that. <g>

                                                      As to value, I have not examined this like you have, so I can't help with that very much, either. When looking for bargains, you should factor in that they are bargains for a reason, and there will always be risks and costs. There is no way to entirely remove either.

                                                      Ron Wodaski



                                                      On Mar 15, 2013, at 10:03 AM, Orlando Andico <orly.andico@...> wrote:

                                                      > Ron, this specific camera already has a shutter driver. 24V trigger, 4V
                                                      > hold.
                                                      >
                                                      > So given that a new 6303 camera starts at about $7K, what the liquidator is
                                                      > asking ($1K) is reasonable? even when factoring in the machining work to
                                                      > attach a shutter and the cost of MaximDL (although MaximDL is not
                                                      > required.. it comes with software that can capture, then just use PHD to
                                                      > guide).
                                                      >
                                                      >
                                                      >
                                                      > On Sat, Mar 16, 2013 at 1:00 AM, Ron Wodaski <yahoo@...> wrote:
                                                      >
                                                      >> **
                                                      >>
                                                      >>
                                                      >> You can roll your own shutter with parts from Uniblitz (you need a
                                                      >> shutter, and a driver to trigger it). Not exactly cheap, but workable. This
                                                      >> assumes the camera has a relay connection to trigger an external shutter;
                                                      >> many do.
                                                      >>
                                                      >> Ron Wodaski
                                                      >>
                                                      >
                                                      >
                                                      > [Non-text portions of this message have been removed]
                                                      >
                                                      >
                                                      >
                                                      > ------------------------------------
                                                      >
                                                      >
                                                      >
                                                      >
                                                    • Stan
                                                      ... That s pretty bright. BTW, you can also use my method/calculator for sky brightness (can be used for selected spectro, i.e. calculate sky brightness for
                                                      Message 26 of 29 , Mar 15, 2013
                                                      • 0 Attachment
                                                        --- "Orly" <orly.andico@...> wrote:
                                                        > I just measured my NELM...
                                                        > is 16.87 mag / sq arc-sec.

                                                        That's pretty bright. BTW, you can also use my method/calculator for sky brightness (can be used for selected spectro, i.e. calculate sky brightness for filter pass-band):

                                                        http://www.stanmooreastro.com/CCD_topics.html
                                                        (second link down)

                                                        > given such a (terrible) amount of sky fog, is my goal still
                                                        > practical?

                                                        what exactly is your goal?

                                                        If it is emission nebula then use narrow band filters to remove nearly all sky glow. The limiting noise becomes readout and dark current (not sky).

                                                        If you want galaxies then go astro-camping! (great fun)

                                                        Stan
                                                      • jtorelli76063
                                                        I missed the beginning of this thread. So excuse me if it is out of context. I have a StarShoot Pro v2 and it does not have a shutter. I use my flip mirror as
                                                        Message 27 of 29 , Mar 17, 2013
                                                        • 0 Attachment
                                                          I missed the beginning of this thread. So excuse me if it is out of context.
                                                          I have a StarShoot Pro v2 and it does not have a shutter. I use my flip mirror as my shutter. I connected a servo motor with a belt and gears to the knob of the flip mirror. I found a serial ascom filter wheel driver on the web. I programmed a small micro controller to talk the ascom driver.
                                                          Now in Maxim I check "use filter as shutter". And maxim will operate the flip mirror when it needs to close the shutter. I also added a switch on the micro so I can still use the Flip Mirror as a flip mirror. I had to make a small aluminum bracket to mount the motor. I used existing holes. I can post photo's and code if anybody is interested.

                                                          JoeT
                                                          BlindEye Obs

                                                          --- In ccd-newastro@yahoogroups.com, Ron Wodaski <yahoo@...> wrote:
                                                          >
                                                          > You can roll your own shutter with parts from Uniblitz (you need a shutter, and a driver to trigger it). Not exactly cheap, but workable. This assumes the camera has a relay connection to trigger an external shutter; many do.
                                                          >
                                                          > Ron Wodaski
                                                          >
                                                          >
                                                          >
                                                          > On Mar 15, 2013, at 9:09 AM, Stan <stan_ccd@...> wrote:
                                                          >
                                                          > > --- Orlando Andico <orly.andico@> wrote:
                                                          > >> ... or are they all more-or-less equivalent?
                                                          > >
                                                          > > Essentially all cameras based on any particular CCD are equivalent in terms of QE, read noise and dark current (per temperature) as there is little that a camera manufacturer can do about those attributes other than screw them up. The things that differentiates such cameras are form factors (e.g. lab cameras are usually large and heavy), built-in features (many lab cameras lack an internal shutter, which can be very problematic for flat fields), cooling, lens mount (usually c-mount or proprietary for lab cameras), data transmission interface (and speed of download), software drivers and compatibilities. There are also tangential but potentially important issues such as repair costs (most lab cameras cost more to repair than a new amateur camera) and obsolescence (e.g. 32 or 32/64 PCI slots have become very rare in motherboards and that card will not work in a 64 PCI bit slot).
                                                          > >
                                                          > > If you are OK using an old desktop machine for the camera and can figure out some sort of robust shutter for flat fields then it might be worth pursuing. The included (or more likely downloadable) software should be sufficient for manual exposures (MaxIm could be used for semi-automation, though if you don't already have it, it is a significant added expense). But it is not for the weak... <g>
                                                          > >
                                                          > > Comparing different CCDs is more complicated because pixel size and noise characteristics have differing effects for particular scopes, filters, applications and techniques.
                                                          > >
                                                          > > Stan
                                                          > >
                                                          > >
                                                          > >
                                                          > >
                                                          > > ------------------------------------
                                                          > >
                                                          > >
                                                          > >
                                                          > >
                                                          >
                                                        • Stan
                                                          ... Do you take/use flats? A flip mirror probably produces bad flats because it exposes one side longer than the other, which would produce a gradient. I
                                                          Message 28 of 29 , Mar 18, 2013
                                                          • 0 Attachment
                                                            --- "jtorelli76063" <josephtorelli@...> wrote:
                                                            > I use my flip mirror as my shutter...

                                                            Do you take/use flats?

                                                            A flip mirror probably produces bad flats because it exposes one side longer than the other, which would produce a gradient. I suppose it might be a small gradient if the flip was fast and the flat exps were long (dim target). And if it is highly stable and repeatable then you could calculate the gradient and remove it mathematically.

                                                            Stan
                                                          • Stan
                                                            ... I just realized that is probably an interline CCD and does not need a physical shutter other than as a convenience for darks (a lens cap would do). But for
                                                            Message 29 of 29 , Mar 18, 2013
                                                            • 0 Attachment
                                                              --- "jtorelli76063" <josephtorelli@...> wrote:
                                                              > I have a StarShoot Pro v2 and it does not have a shutter.

                                                              I just realized that is probably an interline CCD and does not need a physical shutter other than as a convenience for darks (a lens cap would do).

                                                              But for non-interline (or Frame Transfer) CCDs a fast and even physical shutter is important.

                                                              Stan
                                                            Your message has been successfully submitted and would be delivered to recipients shortly.