So why isn't aperture important for DSO's?

[biondi]

Hi,

Often I see in this forum discussion about telescopes and good set-up's for DSO AP. However, I don't understand why the aperture of the OTA isn't really important here? Surely the more light coming in the better?

[Jessun]

Well it is, but it's hidden away in the f-ratio discussion for imagers.

You can determine a sampling scale, which is arc seconds per pixel on your CCD. Call it resolution if you like. This is determined by the pixel size of your CCD and focal length only.

Once that is done it's down to how fast you want to fill the CCD with signal. By making the aperture bigger - keeping the focal length fixed - the f-ratio decreases to a smaller number, and by that the exposure times decrease drastically, meaning that you collect more photons per time unit, due to the bigger aperture. The resolution remains the same.

Time is of the essence, you want loads of data, and quickly. So the aperture is there, as a factor just as vital as any, but it's the f-ratio that gives an imager a better clue of the performance of a telescope, so discussions tend to stay there.

Sadly, large lenses are difficult to make and can budge under their own weight, and large mirrors are also very expensive. Weight plays a big role for what we can possibly carry and set up, and the mount has it's limits too. So again the big telescopes are out of reach for most and we settle for smaller. Then it's a race to make the small telescope as 'fast' as possible - at a premium price too....

Since we expose for hours on end we can still pick up faint stuff that the naked eye looking through the same telescope couldn't.

/Jesper

Edited August 30, 2013 by Jessun

[VigdisVZ]

It's not "unimportant". It's however much less important than mounting.

You can see stunning pictures made with small lens APO's, but you rarely see those long-sub, narrowband composites executed well on a flimsy mount because you'll be throwing away too much material when the mounting is sub-par.

Also, aperture works somewhat in reverese: you will have more strain on the mount, and the focal length might make it impractical since some objects wont fit. I have also heard that large aperture photo-newts take a lot of TLC to get work properly with collimation, and I suspect that larger apertures generally might be more demanding on maintenance.

I tend to think of it as long-distance vs sprinting. When visually observing you want large aperture since the eye cannot alter its "exposure time" and basicly cram as much light in there as possible. However when you're set for a long distance run, you want to go slow and steady to to get a consistent pace and should only go faster if you know can last for the entire race.

[DrRobin]

Hi,

During typing the reply, Jesper and Carl both made several of the points here, they beat me to it.

It is and it isn't, it all depends on how you look at it. You are right as you increase the aperture you will gather more light from a star and hence aperture is king. However, a DSO isn't a single star but a series of stars spread over a field of view and for AP the exposure time is more to do with the F number than the aperture.

For two scopes with the same focal length, lets say 1000mm, a 200mm aperture will have be F/5 where as a 300mm will be F/3.3. The difference in light they will gather/exposure is the square of the F number ratio or 2.29x in this case. You will note that as these scopes are the same focal length then the ratio is just equal to the aperture, so aperture is really king.

But it isn't. Imagine a comparison between my 200mm F/5 scope (1000mm focal length) and a 300mm F/10 scope (3000mm focal length). Comparing aperture should tell you that that the 300mm scope will gather 2.29x as much light so the exposures should be shorter (just over twice as quick). However, comparing the F ratio actually shows that the that the 300mm scope is actually twice as slow and exposures will take twice as long!

How can this be, the 300mm scope has a much bigger aperture, physical laws are breaking down here? Not so, the reason is the focal length has gone up from 1000mm to 3000mm, so it's like trying to compare apples with oranges, if we just look at the aperture. The 300mm scope will have a much smaller FoV and therefore higher magnification, but the exposures will be twice as long.

So, for DSO AP, go on the F number, it is a lot closer than trying to compare apertures. For point sources like stars, the F number doesn't really work since increasing the magnification doesn't result in any bigger star, so for just a single star the exposure times are somewhere between the aperture ratio and the F number ratio.

I hope this answers your question.

Robin

Edited August 30, 2013 by DrRobin

[biondi]

Jesper / Carl / Robin,

Thanks so much, perfect.

[AlexB67]

Whilst I am not an imager of any kind. A simple way of looking at it is that in photography you have one extra variable under your control to collect light and that is exposure time, it feeds into the above and everything that has already been said. For visual observing you do not have the ability to do that, whatever amount of light falls on the eye depends totally on the aperture or light collecting of the scope, but in AP you can play with both, therefore aperture is less critical in total.

Edited August 30, 2013 by AlexB67

[DrRobin]

The F number is the same for visual as well as AP. A system with a lower F number will show objects that are brighter. The other thing we can't control with the eye is image scale. If an object is just too small in the eyepiece to see any detail then you have to increase the magnification and the object will get dimmer as you change the focal ratio of the scope by changing eyepiece. If it was on the edge of visibility then in a lower powered eyepiece then it will go altogether in the higher magnification.

Robin

[JamesF]

The Hyperstar rig on a large SCT is an interesting illustration of some of the points above regarding f-ratio. The characteristics of the OTA are massively altered when swapping between the standard configuration and the Hyperstar setup, without changing the aperture at all. It also shows up some of the difficulties with having exceptionally fast f-ratios, such as razor-edge focusing

James

[StuartJPP]

I come from a purely photographic background where aperture is everything (to a degree).

Since I use a camera lens for my astrophotography that still applies i.e. a 500mm f/4 lens is twice as fast as a 500mm f/5.6 lens. The focal length is fixed at 500mm for both so the field of view (FOV) is exactly the same however I only have to expose for ½ the duration with the f/4 lens than I would have to with the f/5.6 lens. That is the only benefit of course, but for me it is a very important one since I don't guide.

It is the focal ratio that makes a difference and as mentioned above, the Fastar/HyperStar systems change the focal ratio by reducing the focal length (i.e. the light is reflected off the primary mirror straight into the imager).

When I first read the title "Aperture doesn't mean diddly squat" or something similar to that, I thought what a load of rubbish, but now I (hopefully) understand where it is coming from.

[Cath]

Another advantage of larger aperture is increased resolution over using a smaller aperture, at least as far as I know.

[ollypenrice]

I come from a purely photographic background where aperture is everything (to a degree).

Since I use a camera lens for my astrophotography that still applies i.e. a 500mm f/4 lens is twice as fast as a 500mm f/5.6 lens. The focal length is fixed at 500mm for both so the field of view (FOV) is exactly the same however I only have to expose for ½ the duration with the f/4 lens than I would have to with the f/5.6 lens. That is the only benefit of course, but for me it is a very important one since I don't guide.

It is the focal ratio that makes a difference and as mentioned above, the Fastar/HyperStar systems change the focal ratio by reducing the focal length (i.e. the light is reflected off the primary mirror straight into the imager).

When I first read the title "Aperture doesn't mean diddly squat" or something similar to that, I thought what a load of rubbish, but now I (hopefully) understand where it is coming from.

I think that, in effect, daytime photographers use 'aperture' as a synonym for F ratio. The aperture control on a camera does alter the aperture (and so the F ratio) by stopping down internally, but (and herein lies the confusion) the aperture controls of the camera are specified in F stops, which are F ratios. They are not measurements of aperture. A measurement of aperture would be the mm diameter of the diaphragm. There is a perfectly good reason for this misuse of language ( ) because calibrating 'aperture' in F ratio units means that the photographer is using the terms which are useful. He does'nt care how big the diaphragm aperture is. What he needs to know is the photographic speed and depth of field that this aperture creates. (For astrophographers all is at infinity so depth of field doesn't much matter, though the shallower it is the harder it is to get everything to work, notably regarding focus.)

But it cannot be formally correct to say, 'The aperture of this system is F4.8.' The F ratio of the system is 4.8 and the aperture is the diameter of the incident beam. Common usage in photography, though, does not have it that way. However, nobody would ever describe the aperture of their telescope as F5.

They're a stubborn lot, these daytime photographers, but in astrophotgraphy this clash of calibration terminology sometimes causes confusion.

I've read many answers to questions such as that posed by the OP but never read a better answer than Jesper's.

Cath's right that increases in aperture do theoretically bring increases in resolution but within the ranges applicable in amateur AP the effect is fairly slight and can disappear if the larger/faster system suffers from the ills which sometimes do affect very fast systems. An 8 inch Hyperstar should out resolve a 4 inch Takahashi FSQ but would you put money on it doing so for real? I wouldn't.

Olly

Edited August 30, 2013 by ollypenrice

[biondi]

Cheers Olly, this thread has been really useful for me..

[dph1nm]

Oh go on, I will mention the 'f-ratio myth'. Look it up on Google ...

... and just reflect on the fact that professional astronomers build ever larger aperture telescopes for a good reason.

NigelM

[Jessun]

We don't have the budget of professional astronomers .

Very few segmented mirrors come up on AstroBuy, hardly any guide star lasers and just very vew active optics systems. Interferometric imaging kits are even more rare .

We're stuck in the amateur section of things, and resolution down to what seeing allows is for many of us easily reached with a DSLR and an APO small enough to carry in one hand. And I know I'd prefer a fast one of those APOs.

It's all linked of course, change one parameter and another one will change too. For a given focal lenght the only way to decrease f-ratio is to up the aperture, and while it's true that aperture determines the faintest object you can resolve, chances are that the pixels of your average CCD will set the hard limit for that anyhow, so it would make sense to talk about how to fill those pixels up as fast as possible. (I don't talk about planetary imaging here, where tiny pixels and a whole different data capture process is used).

I think this is the reason aperture alone isn't really mentioned all too often in discussions - despite the inevitable link to all other parameters. A little Takahashi at f3.9-something is a bigger buzz these days.

/Jesper

[Big Jim Slade]

I think that, in effect, daytime photographers use 'aperture' as a synonym for F ratio. The aperture control on a camera does alter the aperture (and so the F ratio) by stopping down internally, but (and herein lies the confusion) the aperture controls of the camera are specified in F stops, which are F ratios. They are not measurements of aperture. A measurement of aperture would be the mm diameter of the diaphragm. There is a perfectly good reason for this misuse of language ( ) because calibrating 'aperture' in F ratio units means that the photographer is using the terms which are useful. He does'nt care how big the diaphragm aperture is. What he needs to know is the photographic speed and depth of field that this aperture creates. (For astrophographers all is at infinity so depth of field doesn't much matter, though the shallower it is the harder it is to get everything to work, notably regarding focus.)

And, of course, the whole F-stop language is deeply tried into the carefully established reciprocal relationship between aperture, shutter speed and film speed - all of which helps a photographer work creatively.

Interestingly enough, if you're a cinematographer, it's a bit more realistic. Cine lenses are calibrated in T-stops, which are based on the actual amount of light passing through the lens (on a test bench) rather than the F ratio. This means you can swap lenses between shots and be sure your lighting setup will look exactly the same when you stop down to (say) T4, whichever lens is on the camera.

[ollypenrice]

Oh go on, I will mention the 'f-ratio myth'. Look it up on Google ...

... and just reflect on the fact that professional astronomers build ever larger aperture telescopes for a good reason.

NigelM

There is no F ratio myth if you refuse to compare two systems of different focal length when discussing the merits of aperture. If you respect this rule you are, in effect, talking only about F ratio and then we are back where we started.

I perceive the F ratio myth as 'the focal length myth.' You cannot just rescale a short focal length image to the scale equivalent to that of a longer FL image and expect it to compare. But does anybody really believe it would? I suspect that the F ratio myth is a myth!!! (Not because it isn't true but because nobody every really claims that it is...)

Olly

Edited September 2, 2013 by ollypenrice

[wesdon1]

On 30/08/2013 at 12:50, Jessun said:

Well it is, but it's hidden away in the f-ratio discussion for imagers.

You can determine a sampling scale, which is arc seconds per pixel on your CCD. Call it resolution if you like. This is determined by the pixel size of your CCD and focal length only.

Once that is done it's down to how fast you want to fill the CCD with signal. By making the aperture bigger - keeping the focal length fixed - the f-ratio decreases to a smaller number, and by that the exposure times decrease drastically, meaning that you collect more photons per time unit, due to the bigger aperture. The resolution remains the same.

Time is of the essence, you want loads of data, and quickly. So the aperture is there, as a factor just as vital as any, but it's the f-ratio that gives an imager a better clue of the performance of a telescope, so discussions tend to stay there.

Sadly, large lenses are difficult to make and can budge under their own weight, and large mirrors are also very expensive. Weight plays a big role for what we can possibly carry and set up, and the mount has it's limits too. So again the big telescopes are out of reach for most and we settle for smaller. Then it's a race to make the small telescope as 'fast' as possible - at a premium price too....

Since we expose for hours on end we can still pick up faint stuff that the naked eye looking through the same telescope couldn't.

/Jesper

You just answered the exact question that's been bugging me for months lol. You have made perfect sense of it for me,  even though i didn't actually ask you! Biondi did! So thanks to you too Biondi for asking the question we both wanted to know! lol I'm only 7 months into hobby so any and all knowledge gained is always greatly appreciated. Thanks guys.

[Jessun]

Wow, what a blast from the past!

I'm happy that it cleared up some issues. The basics of optics is never determined by one parameter alone, they are all linked.

Think about this: The Hubble Space Telescope is as far as I can gather f24.

The focal length of the HST is something like 60 meters vs a 2.5m mirror, give or take. You can build the same f-ratio telescope at home, scaled down to a few lengths of toilet rolls and if you achieve same f-ratio, (A two inch toilet roll would have to be stacked to a length 24 times the aperture  meaning a 48 inch telescope for that tiny aperture). 

If you built this and put it in orbit, or indeed kept it in the back yard, it would fill the wells of a given CCD or CMOS at the same rate as the Hubble. OK, perhaps only for the very central pixels as the light cone fades rapidly towards the edges, but the f-ratio does not take this into the equation...

When that fundamental penny drops, there is no longer a mystery concerning f-ratio ever again.

/Jessun

Edited July 24, 2019 by Jessun

[DrRobin]

Simply put.....

F ratio determines how long you will image for (your exposure time);

Aperture (and focal length) determine how much you will fit in.

 

E.g. a 12" F/5 telescope will have the same exposure as a 6" F/5 telescope, its just that the 6" will cover a much wider field.

[dph1nm]

18 hours ago, Jessun said:

f you built this and put it in orbit, or indeed kept it in the back yard, it would fill the wells of a given CCD or CMOS at the same rate as the Hubble

Sadly you wouldn't see many objects (in the same exposure time as a typical Hubble shot), as how faint you can see depends on the aperture not the focal ratio.

NigelM

[DrRobin]

10 minutes ago, dph1nm said:

Sadly you wouldn't see many objects (in the same exposure time as a typical Hubble shot), as how faint you can see depends on the aperture not the focal ratio.

How faint you can see depends on how many arc-seconds per pixel and the signal to noise ratio of the chip.  Hubble is in space so has very low noise.  The F-ratio, aperture and pixel size all then contribute to the number of photons in each pixel from any given light source.

If you increase aperture and as a consequence increase F/ratio (focal length) then the number of photons per pixel might not change and you won't get any more signal.  If you increase pixel size then you will see faint objects provided you don't increase noise as a result, but at the expense of resolution.

[vlaiv]

5 hours ago, DrRobin said:

Simply put.....

F ratio determines how long you will image for (your exposure time);

Aperture (and focal length) determine how much you will fit in.

 

E.g. a 12" F/5 telescope will have the same exposure as a 6" F/5 telescope, its just that the 6" will cover a much wider field.

Simply put - this is wrong.

Many people forget pixel size - or assume unchanging pixel size. F/ratio does not determine "speed" of setup - meaning it does not determine how long you will expose.

Imagine 8" F/10 scope and 6" F/5 scope for comparison.

First one is used with ~11um pixel camera. Second one is used with ~4um pixel camera. I added ~ sign (meaning about) because I wanted to emphasize that in both cases you have around 1.15"/px sampling rate.

Now you have 8" of light gathering vs 6" of light gathering - both mapping 1.15" of sky to size of pixel. 8" will win on speed, even if F/10 is "slower" scope than F/5 because it is larger aperture.

I agree that you will have wider potential field with shorter focal length within limits of optical design (fully illuminated and corrected circle).

[DrRobin]

2 minutes ago, vlaiv said:

Simply put - this is wrong.

Many people forget pixel size - or assume unchanging pixel size. F/ratio does not determine "speed" of setup - meaning it does not determine how long you will expose.

I was making the assumption that it would be the same camera on both.  If you change the pixel size then everything changes, oh and you might as well change location to above the atmosphere where there is less loss.

5 minutes ago, vlaiv said:

Now you have 8" of light gathering vs 6" of light gathering - both mapping 1.15" of sky to size of pixel. 8" will win on speed, even if F/10 is "slower" scope than F/5 because it is larger aperture.

I doubt this is true either, both systems end up with the same amount of sky per pixel, if both cameras have the same sensitivity (difficult to achieve) then both will image in the same time as they both have the same number of photons to play with.

[vlaiv]

8 minutes ago, DrRobin said:

I was making the assumption that it would be the same camera on both.  If you change the pixel size then everything changes, oh and you might as well change location to above the atmosphere where there is less loss.

One does not need to change the camera, you can bin your pixels to get larger collecting surface / lower sampling rate.

8 minutes ago, DrRobin said:

I doubt this is true either, both systems end up with the same amount of sky per pixel, if both cameras have the same sensitivity (difficult to achieve) then both will image in the same time as they both have the same number of photons to play with.

Don't have to be doubtful about it - do the math and you will see.

Here is very simple example that is enough to show this. Let's say that we have two scopes - one with 8" and one with 6" aperture. We match them with sensors / pixel sizes, so that both give 1"/px. This means that one pixel will gather all photons from 1"x1" patch of the sky that were collected with respective aperture and focused on that pixel (all photons fall on aperture as parallel rays and they are then focused on that pixel so all photons from 1"x1" region that aperture collects end up on that given pixel and are accumulated as signal). Next step is fairly easy one - 8" will collect more photons than 6" -> Pixel on camera attached to 8" scope will gather more photons and have stronger signal than Pixel on camera attached to 6" scope.

Better SNR in same time = same SNR in shorter time.

This is why I always say - don't think in terms of F/ratio or speed of scope, but rather thing in terms of "aperture at resolution". More aperture on same resolution will be faster. That is why are larger scopes better for DSO imaging, or rather primarily reason if you can pair them up with sensor to give you wanted / reasonable resolution (possible binning included in consideration).

Edited July 25, 2019 by vlaiv
typo ...

[DrRobin]

Binning doesn't help that much.  If you use a 2x2 bin (4 times the area), but they way the signal is read it is more like 2x the signal.  To get back to 4x signal you have to use 4x4 bin, roughly speaking.

Smaller pixel sizes often have a lower light sensitive area to total area, due to the need for readout registers.  It's so dfficult to compare different ccds.

If you look at my post from 2013 you will see I differentiated between a DSO (the subject of this thread) and a star.  This is important if you are considering point sources or a light spread out of an area.