The Importance of Aperture in Astrophotography?

[Captain Scarlet]

I was more thinking of the etendue inside the optical system from the aperture to the pixel: ie the angle of view to the pixel from the pov of the aperture, integrated across the aperture. It should be, using those easily known numbers, that two images of the same target using hugely different optical trains but matching the eventual etendue.sqrt(exposuretime) are indistinguishable from each other, image-quality-wise? Apologies if I've de-railed from the OT...

Magnus

[andrew s]

6 minutes ago, Captain Magenta said:

I was more thinking of the etendue inside the optical system from the aperture to the pixel: ie the angle of view to the pixel from the pov of the aperture, integrated across the aperture. It should be, using those easily known numbers, that two images of the same target using hugely different optical trains but matching the eventual etendue.sqrt(exposuretime) are indistinguishable from each other, image-quality-wise? Apologies if I've de-railed from the OT...

Magnus

Not sure I fully understand you. However, I think that you just end up with the aperture being key. As you change focal length the solid angle of the sensor changes but so does the projected area of the sky. If two systems capture the image at the same resolution ( I.e. arc secs  ) only aperture determines  the signal level per arc sec squared. 

Regards Andrew 

[Captain Scarlet]

I think my point is, having read the OP again to try to make it relevant, that the etendue.sqrtT => SNR formula, using internal system values like pixel size, aperture, focal length, is the means by which you, with the same target and the same camera but different telescopes such as 200p or 130pds, can scale exposure time such that you get the same quality image for either...

It's such a shame one can't just drop round to your place/s or arrange to meet and talk about these things, would be so much easier...

Cheers, Magnus

Edited May 24, 2020 by Captain Magenta

[vlaiv]

Just had a thought that could be interesting for discussion on this thread.

We talk about speed of telescope and speed of camera - but those things don't have a speed. They can't produce image on their own. They need to be paired with the other component to be able to produce image.

Thus we can only talk about potential of telescope to be fast and potential of camera to be fast.

In that sense, I feel that aperture is essential component of telescope potential to be fast. Other component that is often overlooked in these discussions would be - fully illuminated and corrected field (well, it does not need to be fully illuminated to be useful, but we do like fully corrected field).

Scope with small useful field has lower potential than scope with large useful field - as later can be paired with larger cameras.

Third component that is neither "good" nor "bad" would be focal length. I think that three of those in some way represent scope's potential to be fast. I still don't clearly see relationship between the three and potential to be fast in a scope.

Anyone has any ideas about this?

[Whistlin Bob]

Thank you very much everyone for trying to help me understand this topic. I've had a go at the maths of this and tried to estimate how much aperture is illuminating each square arcsecond of sky that the camera is imaging- although not all sure I have it right (may just be demonstrating that I haven't followed this properly!!!).

Aperture	FL	Pixel Size	ArcSec/Pixel	ArcSec2/Px	Area in mm2	mm2 / ArcSec2
200	900	3.8	0.87	0.76	125664.00	166024.57
130	585	3.8	1.33	1.77	53093.04	30014.72

I've taken the characteristics of the scope/cc/cameras that I'm using. The column ArcSec2 is meant to be the area of sky each pixel is covering and the Area in mm2 is the area of the mirror, ignoring the central obstruction which is assumed to be proportional for the 2 f5 scopes I'm considering. The final column is then area per square arcsecond arriving on that pixel which is showing a massive advantage for the 200mm scope. 

I'm then looking at @tooth_dr's very helpful images (which is just the sort of experiment that I was lamenting not having done)- I think the 250mm is better, but nowhere near the extent the numbers above would suggest. I appreciate that they are processed, which probably hides quite a bit, but it really shows that improvements are marginal...

Edited May 24, 2020 by Whistlin Bob
type

[tooth_dr]

I don’t know whether the h alpha made much difference? 

[andrew s]

2 hours ago, vlaiv said:

Just had a thought that could be interesting for discussion on this thread.

We talk about speed of telescope and speed of camera - but those things don't have a speed. They can't produce image on their own. They need to be paired with the other component to be able to produce image.

Thus we can only talk about potential of telescope to be fast and potential of camera to be fast.

In that sense, I feel that aperture is essential component of telescope potential to be fast. Other component that is often overlooked in these discussions would be - fully illuminated and corrected field (well, it does not need to be fully illuminated to be useful, but we do like fully corrected field).

Scope with small useful field has lower potential than scope with large useful field - as later can be paired with larger cameras.

Third component that is neither "good" nor "bad" would be focal length. I think that three of those in some way represent scope's potential to be fast. I still don't clearly see relationship between the three and potential to be fast in a scope.

Anyone has any ideas about this?

I think that is what etendue encapsulates.  Am I missing something? 

Regards Andrew 

[vlaiv]

Just now, andrew s said:

I think that is what etendue encapsulates.  Am I missing something? 

Regards Andrew 

Not sure - will need to look it up. I've heard of etendue but have no idea what it is

 

[alacant]

On 23/05/2020 at 15:35, Whistlin Bob said:

hasn't really been able to see much difference between them on individual subs

Hi

I'm surprised at that. Maybe he changed camera? There's a noticeable difference in brightness even when moving from 80mm to 130mm.

Cheers

[vlaiv]

49 minutes ago, andrew s said:

I think that is what etendue encapsulates.  Am I missing something? 

Regards Andrew 

I think that etendue encapsulates much more but also it missis a bit that is important when we want to understand all of this. It is essential to this discussion and I believe pretty much all will agree that it is implicitly implied.

It is essential component of why telescopes work the way they work - conservation law that explains why we see equal brightness of the star regardless where in the fully illuminated field we put it.

However - I believe it is concept that is missing noise part and hence can't be used to explain SNR and consequently speed of the system. It is also very difficult concept to grasp for most people. For this reason, I don't find it very useful in explaining things.

We need to find, or attempt to find something that is easy to understand but does provide accurate and applicable way for people to think when thinking about speed of telescopes. There might not be such thing - at least not easy one.

What I wanted to point was this:

If someone would approach majority of astrophotographers with following question:

ED80 + ATIK 414EX color or ATIK 428Ex color

vs

Skywatcher Maksutov 102mm and ASI294

I have sneaky suspicion that 95% of answers would be along the line - ED80 + atik of course with screams in horror when Maksutov F/13 scope is mentioned in astrophotographic context.

Then I say, let's look at following:

Mak has enough illuminated field to capture same FOV, it has enough aperture advantage to offset mirror reflectivity and central obstruction and will in principle produce same image for same amount of time (at half the price?).

I do agree that Mak is going to be a bit harder to work with - does the mirror shift? You'll need OAG rather than guide scope and so on, but that is besides the point.

But if all I said was - I have 100mm of aperture that covers 0.85 x 0.6 degrees, is it going to be fast - I might get different answer than above.

 

[andrew s]

I fully agree @vlaiv that most would dismiss the Mak however, what you describe in the example is exactly what  I believe Etendue tells you. I.e. Area of aperture x Area of field of view, is the throughput and the Mak wins. 

I also agree that it does not cover SNR directly but for conventional imaging you are normally in the photon shot noise dominated region so telescope aperture is dominant. If you have a bright sky then if your targets are extended and polychromatic then I am not sure what you can do except  move to a darker site.

I think we fundamentally agree.

Regards Andrew 

Edited May 24, 2020 by andrew s