Yes that is a workable truth about the aperture and F ratio and exposure time. But like any theory I am sure there are deviations due to other factors and not all chips have exactly the same characteristics but yes I agree it is a workable rule that two 200mm scopes will expose the same brightness for the same time even if different f ratios. So if you want faster exposure times you up aperture. Although I know from an actual test that a Tak ED180 at F2.8 exposed to same brightness an image 2.5X quicker than an F5 FSQ106. Now that is 74mm more aperture but there is also a central obstruction. Not sure the total area of the aperture is over 2.5X. Probably more like 4% more area.
Collimation is not that big a point either for both scopes. F5 Newt can be tricky with a star test but if you use a cheap laser collimator it takes about 4 minutes (I used to have a Vixen R200SS F4 200mm). The RC optics though I had (RCOS 12.5inch) did require several steps including aligning the mirrors rotation-wise and using a Tak collimating scope as it was pretty impossible without but not hard with it ($400 collimating scope). The GSO may be simplified. I would not scare off based on collimation as it is a skill that should be learned anyway.
I am not criticising the GSO which is great bang for your buck. The question was whether to change the F5 200mm Newt for one.
What my main point was imaging scale. F5 200mm will give a wider field of view and that makes everything easier and more pleasing an image to more people. But then a lot prefer to image at a closer up scale so its what type of image you want to produce.
Studying Stan Moore's 10 minute photograph of F3.8 vs F12 I see that the signal to noise is worse on the F12.4 ( as might be expected) but the resolution is only ever so slightly better.
I'm now convinced that it is better for me to go for an F3.5- F4 Coma corrected Newt than a long focus Cass as I will have a much larger field of view available for those larger objects and I will not be losing anything significant in resolution for those smaller objects ( I guess seeing is a factor here also). I can also use a 2" 2X Televue Barlow for F7-F8 should I need to take advantage of really fine seeing .
Thanks for posting that very interesting paper , Peter. CCD is certainly a whole other world from film.
Just a point here. I think that my GSO RC has primary adjustment screws on it. Located at the rear are three hex headed screws 120 degrees apart on the inner mounting area. I am assuming this is either collimation adjustment or is it just the locking nuts to hold the primary in place? If it is the later how would they have maintained optical alignment. Surely this is conducted by way of a spring forcing against the primary. In which case primary collimation is possible???
2) You are assuming the primary is decentered by the factory. I've tested close to a dozen of these scopes and not one had a problem. Also last time I checked an F5 of anything was far more of a bugger to collimate than a F8
3) The point Greg made was the *tube was longer*.
It is patently not,
and will suffer less from wind loading as a result....though I do take the point the longer focal length could be more sensitive to seeing and tracking errors.
Yes, I do. For a given camera (i.e. given pixel size) and for those objects that are just above noise threshold, faster scope delivers more photons per pixel, which will result in a superior S/N. In fact it may be a matter of detecting the said nebulosity or missing it altogether.
Try those superfaint galactic Mandel-Wilson nebulae with your R-C and your STL11K one day
Yes, I do. For a given camera (i.e. given pixel size) and for those objects that are just above noise threshold, faster scope delivers more photons per pixel, which will result in a superior S/N. In fact it may be a matter of detecting the said nebulosity or missing it altogether.
Try those superfaint galactic Mandel-Wilson nebulae with your R-C and your STL11K one day
Sure, but the contribution from the sky background also *increases* as the f-ratio lowers. This lowers the S/N level.
Yes I agree with you Bratislav and I do think that Stan's paper is an
overall guide but it needs another datum to be fully practical although he acknowledges
the higher F ratio scope will give a lower signal to noise ratio.
My experience with a BRC250 at F5 and an RCOS 12.5 at F9 says that you can get a bright image from the BRC in 2.5 hours but the 12.5 at F9 will require more like 6 hours. Yet the BRC has less aperture. But the RCOS was a more zoomed in view of the object so its collecting a smaller slice of the available photons so the light is less compared to the nosie from the CCD itself.
And that is the point, at F9 you are looking at a small slice of something but with the same noise.
I must make an experiment of this some time to satisfy this point more clearly.
12 inch ASA at F3.6 clearly captures more nebula than similar close up images from 12.5 inch F9 RC's this is evidence that Bratislav is correct.
The BRC250 will also pick up faint nebula that other images I have taken of the same object with the 12.5 inch RCOS did even with longer exposure times. Different cameras nothwithstanding.
Greg.
Quote:
Originally Posted by bratislav
Yes, I do. For a given camera (i.e. given pixel size) and for those objects that are just above noise threshold, faster scope delivers more photons per pixel, which will result in a superior S/N. In fact it may be a matter of detecting the said nebulosity or missing it altogether.
Try those superfaint galactic Mandel-Wilson nebulae with your R-C and your STL11K one day
If you vary the aperture and keep focal length constant, then S/N ratio is a function of the f/ratio.
However the point to Stan Moore's article is what happens when you vary the focal length and keep the aperture constant. In this case S/N ratio is not a function of f/ratio.
S/N ratio is a function of the aperture or area of the primary mirror or lens. For example a 10" mirror at f/5 and a 10" mirror at f/10 will both collect the same number of photons, the only difference being the photons reach focus at different distances from the mirror.
Note that this does not apply when imaging using eyepiece projection. The S/N ratio is a function of the diameter of exit pupil.
The diameter of the exit pupil = focal length of eyepiece/ F-ratio.
However the point to Stan Moore's article is what happens when you vary the focal length and keep the aperture constant. In this case S/N ratio is not a function of f/ratio.
S/N ratio is a function of the aperture or area of the primary mirror or lens.
Steven
Clear an succinct. I get distracted by the details. Bless you Steven...
What practical effect is there then in having 4 times the # of photons hitting a CCD pixel per unit of time given two telescopes of given aperture , same camera and one at F8 and one at F4?
Quote:
Originally Posted by sjastro
F-ratio = Focal Length/Aperture.
If you vary the aperture and keep focal length constant, then S/N ratio is a function of the f/ratio.
However the point to Stan Moore's article is what happens when you vary the focal length and keep the aperture constant. In this case S/N ratio is not a function of f/ratio.
S/N ratio is a function of the aperture or area of the primary mirror or lens. For example a 10" mirror at f/5 and a 10" mirror at f/10 will both collect the same number of photons, the only difference being the photons reach focus at different distances from the mirror.
Note that this does not apply when imaging using eyepiece projection. The S/N ratio is a function of the diameter of exit pupil.
The diameter of the exit pupil = focal length of eyepiece/ F-ratio.
What practical effect is there then in having 4 times the # of photons hitting a CCD pixel per unit of time given two telescopes of given aperture , same camera and one at F8 and one at F4?
The number of photons hitting a given pixel is determined by
(1) The brightness of the object being imaged.
(2) The aperture of the telescope.
(3) The exposure time.
All things being equal if f-ratio is altered by focal length, the pixel scale is the only factor that is changed. This leads to a change in the size of the field of view.
...........All things being equal if f-ratio is altered by focal length, the pixel scale is the only factor that is changed. ..........
Steven
Steve..would a rain gauge analogy be a good one? vis
x amount of rain falls from the sky (aperture), and we measure it with a rain gauge (small pixel), and say a swimming pool (large pixel)...in either case the level goes up by only x mm.
Steve..would a rain gauge analogy be a good one? vis
x amount of rain falls from the sky (aperture), and we measure it with a rain gauge (small pixel), and say a swimming pool (large pixel)...in either case the level goes up by only x mm.
I think the analogy is fraught with danger. The rain gauge can overfill (equivalent to saturation) and cause all sorts of complications.
The point is if you reduce the f/ratio by reducing the FL, and keep the exposure and aperture equivalent, you are not adding to the total signal hence the noise in the image (the statistical variation of the signal) is the same irrespective of the f/ratio.
The photons will fall over a smaller number of pixels and while it is true that any given pixel will experience an X-fold increase in the number of photon hits, the noise per pixel increases by the same factor.
So as the raw image projected on the CCD chip is 4X as bright on the F4 and an F8 scope, won't the brightest parts of the object fill the CCD pixel to full well capacity much faster . So if you can get more exposures in doesn't more stacked images that are at full saturation mean better signal /noise ?
I'm on a learning curve here , but this is one hurdle I'm not over yet...
Sorry...I had to think about why this was incorrect.....
This is the same as saying: increasing the pixel size will increase the S/N.
Bollocks (haven't said that for a while )
System gain indeed goes up, but the S/N does not.
You can be as colorful as you want; that doesn't change facts. There is more that comes from a pixel than a simple background noise. You also have photon noise (or sometimes called shot noise), dark noise and readout noise. These are factors even in liquid nitrogen cooled front illuminated scientific CCDs, let alone consumer units we deal with. Those are constant for a given exposure (some will go up with exposure, some won't, like quantization noise or output amplifier or 1/f noise for example). So if you expose faint nebula for X minutes, faster scope will deliver more photons per pixel, you will get more electrons generated in the well and that will result in higher signal to noise ratio.
Stan's analysis is sound if, and only if, you use correspondingly larger pixel camera (with equivalent noise characteristics) on a longer f/ratio scope. But amateurs never do that - we simply use whatever CCD we've got. And in that case, faster scope will deliver higher S/N for extended objects. This advantage does diminish as signal goes up - so for bright objects S/N gets mainly dominated by background noise, and you can stretch the image, as Stan did, to make it look equal.
But for very faint extended objects that may be just above combined background/camera noise, this is not the case.
if increasing pixel size does not increase S/N why would you ever bother binning 2x2? I suppose though, that is not physically increasing your pixel size, rather, adding 4 pixels worth of signal into 1?
)
(haven't said that for a while 
