Ok - ive got analysis paralysis.. SOS! (Which OTA..)

[Slawomir (Suavi)]

My understanding and experience agree with what Colin has written about F-ratios and photographic speed. Large aperture will get you more stars and they will be brighter largely irrespective of the F-ratio, while faster F-ratio will give your subs a greater SNR for extended objects. (This is actually good news when using pixels with shallow wells on a fast small aperture telescope, because stars do not saturate very quickly while we get good signal on extended objects)

For example: 15-minute subs at f/4.5 still had a higher SNR than 20-minute subs at f/6 (same aperture and same camera, same mount etc) - verified on a number of nights when I was going for over 200 hours on Helix.

With a short refractor - personally I would not want to go faster than f/4.5. Focus shift becomes significant and precise orthogonality to ensure a well-corrected field is also becoming a challenge when using small pixels. I also agree that the corrected flat coma-free field shrinks rapidly when we attach stronger reducers.

[The_bluester (Paul)]

That is interesting Suavi, I never really thought of it like that which might be one reason why you see such good results around from combinations like a RASA 8" and ASI1600 which used as most people do, has effectively about 8K pixel wells, so very shallow. At the relatively short focal length the "Point source" of stars is much more like a point source as the airy disc will not be as prominent where the nebulosity will benefit a lot from the F2.

Makes me think when they finally release the reducer/corrector for my SVX80T I should buy it.

[Emuhead (Andrew)]

If we had two 130mm scopes, one at f/10 and one at f/4, focal length aside (and visual astronomy also aside), 30 seconds on the same target would yield very different results right?

Lets say that we even allowed for the focal length and cropped the wide f/4 image down to match the narrow f/10 field-of-view image. Comparing just those 2 resultant images... the f/4 cropped image would have to have collected more photons just in that specific cluster of pixels right...OR...are we saying that number of 'photons per pixel' would remain more or less constant given the aperture is 130mm for both scopes and therefore the cropped area of pixels would essentially have to have collected a fraction of the total photons (relevant to whatever the crop size was).

So for example lets say the crop covers only 1/4 of the sensor, this would then mean that the cropped f/4 image could only have collected 1/4 of the photons as compared to the f/10 image (which used its entire sensor).

After all there's only so many photons you can cram down a tube..and thats probably agnostic of what f/number the scope is...so that constant has to be governed by the aperture, and its whether you spread those collected photons over a wide area of sky, or a narrow area of sky (ie. focal length as determined the by f number).

So does this mean the 120mm will have to collect more photons per second than the 100mm (regardless of f number)...
Edit: When i have that out loud... its seems obvious.

[Slawomir (Suavi)]

Yes, a 120mm aperture collects more photons than a 100mm aperture, but we do not bring all those photons into a single point on a sensor. If we did, our images would be just one bright point at the centre of a sensor, and larger aperture would make that point, or a blob, brighter. Instead, some/most of the light collected by aperture is being distributed over the entire sensor, and it is the f-ratio, not the aperture, that will determine the brightness of extended objects being projected on the sensor. Stars, on the other hand, benefit from large aperture. And so does resolution of images of the extended objects.

[Atmos (Colin)]

It ALL comes down to flux; the amount of photons per unit area of sky. Both telescopes have the same aperture so they both capture the same amount of light. The difference between them is their f/ratios squared.

10^2/4^2 = 6.25
So the F/4 is 6.25x faster than the F/10

To show this, let’s use an ASI1600 with 3.8 micron pixels.
With the F/4 and a FL of 520mm, you have an image scale of 1.51”/pixel.

With the F/10 and 1300mm FL, an image scale of 0.6”/pixel.

Let’s say with a 130mm telescope you capture 100e- per square arcsec of sky per second. That’s the sky flux.
With the F/4 it’s (1.51^2)*100=228 electrons/s per pixel
With the F/10 it’s (0.6^2)*100=36 e-/s/pixel

So you’d have to capture 6.33 (rounding errors) times longer with the F/10 to capture the same amount of electrons.

Another example would be to have a 325mm aperture with 1300mm focal length. It’s a 12.8” F/4 so it captures 6.25x more light than the 130mm. It will also be imaging at 0.6”/pixel but the flux increases 6.25x.
So, (0.6^2)*625=225e-/s/pixel

Remember, 130mm gives a flux of 100e-/arcsec so a 325mm gives 625e-/arcsec.

[Emuhead (Andrew)]

Thanks Colin, so we might as well just use my inputs:

Fuji XT2:
24.3 Megapixels
Sensor size: 23.6mm x 15.6mm
Pixel size = 3.9 microns
Resolution = 6000 x 4000 pixels

Esprit 100ED f/5.5 550mm
Esprit 120ED f/7 840mm

So.. 7^2/5.5^2 = 1.62
So the f/5.5 is 1.62x faster than the f/7
Although I think the above is only usable if the apertures were the same size right. So might have to disregard that for now.


550mm with my sensor gives 1.463"/pixel
840mm with my sensor gives 0.958"/pixel
Thanks to this calculator

Let’s say with a 100mm telescope you capture 100e- per square arcsec of sky per second. That’s the sky flux.
With the 100mm F/5.5 it’s (1.46^2)*100= 213 electrons/s per pixel

Does this mean that with a 120mm, we can capture more light per square arcsecond? If so, given area of 100mm is 7853mm2 and area of 120mm is 11309mm2, then thats 1.44x more area, so 1.44x more ability to collect photons.. im going rogue here, those who know please correct me..

So then with that, let’s say with a 120mm telescope you capture 1.44*100e- per square arcsec of sky per second. That’s the sky flux.
With the 120mm F/7 it’s (0.958^2)*144 = 132 electrons/s per pixel

So you’d have to capture 1.61 (rounding errors) times longer with the 120mm F/7 to capture the same amount of electrons as the 100mm f/5.5.

Is this about right? Or do you actually capture less light per square arcsecond.. my brain hurts. I bet my assumptions are way off.. or ive allowed for something twice.. or not allowed for something.

[FrancoRodriguez (Franco)]

I've got an esprit 100. All around great scope, but... the centre of the image is crystal clear but I'm a little disappointed at the field curvature in the peripheries, even on my ASI071MC APS-C sensor. It does <<almost>> fine on a SX814 sensor. So, great with small sensors, not so great with medium sized sensors. Forget full frame

[kosborn (Kevin)]

I have an Esprit 100 and love it. The flattener works well but spacing to the sensor is critical. It has a tolerance of +/- 1mm I think but outside of that, the stars around the edges of an ASI1600 aren't round. Obviously more so with an APS-C sensor or larger. The FOV on the Esprit works very well for most nebulae but I added a Starizona Apex x0.65 recently. It does a great job of flattening the field with a bit more tolerance than the Sky-Watcher flattener and turns the Esprit 100 into a very fast wide field triplet. Of course I fell down the rabbit hole of astronomy purchases and added a Sidereal f/5 254mm Newtonian so with two OTAs I have 1270mm, 550mm and 357.5mm focal lengths ranging from f/5.5 to f/3.575. I would guess that the Esprit120 with the Starizona reducer would give you the best range of focal lengths for nebulae and larger galaxies.

[Camelopardalis (Dunk)]

Originally Posted by FrancoRodriguez

I've got an esprit 100. All around great scope, but... the centre of the image is crystal clear but I'm a little disappointed at the field curvature in the peripheries, even on my ASI071MC APS-C sensor. It does <<almost>> fine on a SX814 sensor. So, great with small sensors, not so great with medium sized sensors. Forget full frame

Sounds like something is out in your optical train.

Skywatcher quote a 40mm imaging circle and I've found that to be quite faithful to the truth on my scope, with only the far corners of the frame slightly out of round with a full frame Canon 6D.

[Emuhead (Andrew)]

Ok, so coming back full circle on this.. can anyone with the formulas to work this out assist in determining once and for sure.. based on any given sensor (but mine is a Fuji XT2) which scope would capture light faster?

100mm f/5.5
120mm f/7

Id be eternally grateful if it could be worked out, and even more grateful if I could understand it afterwards.

Thanks

[Atmos (Colin)]

Originally Posted by Emuhead

Ok, so coming back full circle on this.. can anyone with the formulas to work this out assist in determining once and for sure.. based on any given sensor (but mine is a Fuji XT2) which scope would capture light faster?

100mm f/5.5
120mm f/7

Id be eternally grateful if it could be worked out, and even more grateful if I could understand it afterwards.

Thanks

7^2/5.5^2=1.6198

So your calculations a few posts up were dead on

The reason I went more complicated with my previous post was to show WHY it’s the F/Ratio that determines photographic speed and not aperture. It can also prove that my 50mm F/1.4 is photographically 25x faster than the 120mm and 15.5x faster than the 100mm

[Emuhead (Andrew)]

The part thats gets me is, how this calculation can be true for different apertures.. i would totally get this if we were talking the same apertures (100mm only), but we have 2 different apertures 100mm & 120mm. I just cant get my head around how that one formula can be good for any 2 different apertures. Say a 50mm and a 1000mm, both were f/2. This calculation breaks here.. because 2^2/2^2 = 1. They both cant catch light at the same speed.. the 1000mm would be so much faster..

[Atmos (Colin)]

Originally Posted by Emuhead

The part thats gets me is, how this calculation can be true for different apertures.. i would totally get this if we were talking the same apertures (100mm only), but we have 2 different apertures 100mm & 120mm. I just cant get my head around how that one formula can be good for any 2 different apertures. Say a 50mm and a 1000mm, both were f/2. This calculation breaks here.. because 2^2/2^2 = 1. They both cant catch light at the same speed.. the 1000mm would be so much faster..

It’s not necessarily about the amount of light that gets into the system. What it is really about is how that light is concentrated and the focal ratio determines how concentrated the light is.

As aperture increases the amount of flux per arcsec of sky increases BUT at the same time the amount of sky that each pixel covers decreases proportionally. Let’s take your Fuji as an example.

A 100mm (4”) F/2 will have an image scale of 4”/pixel so it’s very low resolution. The 1000mm (40”) F/2 will have an image scale of 0.4”/pixel which is quite high resolution. If I do 4^2/0.4^2 (image scale) it shows that each pixel in the 1000mm is covering 100x less area of sky than the 100mm. But the 1000mm is also capturing 100x more photons so it balances itself out.

If a 100mm telescope captures 100e- then my 50mm F/1.8 (35mm aperture) only captures 12.76e-/arcsec/s which is considerably less. The caveat is that each pixel covers 16 arcsec as opposed to the 1.463”/pixel of the 100mm Esprit which is 121x more area! It’s only capturing 7.8x less light though which is why it’s 7.8/121=15.4x faster.

[Emuhead (Andrew)]

Thanks Colin - i get it now.. thats actually a relief!

I have decided to save the extra cash and stick with the 100ED for now too, on account of being cheaper, and concentrating more light on the sensor than the 120ED. Whenever i look at a mono camera down the track, ill probably revisit the 120ED then.

Much appreciated once again.

[The_bluester (Paul)]

If it is your first imaging setup, you will thank yourself for the shorter focal length when it comes to guiding it that is for certain. And the shorter focal length scopes can produce some really nice results. I shot these with my 80mm Stellarvue and ASI294MC Pro camera (Which is quite tricky to calibrate nicely with flats and bordering on undersampled at this focal length) The third is a WIP of a mosaic I am trying to finish this season.



https://www.astrobin.com/full/1bouo7/0
https://www.astrobin.com/full/bcchy3/0
https://www.astrobin.com/full/5qg6os/0

[Emuhead (Andrew)]

So I'm looking directly at the RASA 8 now.

7x faster than the 100ED, and 12x faster than the 120ED.

Can you get nice round coloured stars with this scope? Any glaring reasons not to go the RASA?


Hi Paul, I like the second image a lot, the stars look 3D.. really nice image.

[The_bluester (Paul)]

I think that h0ughy's images there have some of the best looking stars I have seen out of a RASA 8, which proves that it can be done. The trick is that at F2 getting sensor tilt dialled in and the spacing correct it really critical as the CFZ is tiny (Critical Focus Zone, I was trying not to write Critical twice in quick succession, now I have done it three times!)

Get the spacing wrong or the sensor not perpendicular to the focal plain and you will have wonky stars, but the RASA is not alone there. I am still not happy with the corners using my SVX80T, but I have new hardware coming that will change things so settled on "good enough for now"

For something like the RASA I reckon automated focus will be a very good thing to have, though they are probably not as tough to focus as F2 compared to a normal SCT at F6.3 (Reduced) might indicate, and from what I have read they are more focus stable with temperature. The double mirror setup multiplies your pain in the SCT configuration compared to the single mirror RASA both getting it focused and keeping it there. But if you want to be able to leave it imaging all night with sharp subs from end to end automated focus is a great thing, if you like to sleep.

[JA]

Originally Posted by Emuhead

So I'm looking directly at the RASA 8 now.

7x faster than the 100ED, and 12x faster than the 120ED.

Can you get nice round coloured stars with this scope? Any glaring reasons not to go the RASA?

Hi EH,

No- no glaring reasons, but since you appear to basing your decision somewhat on the numbers, and there is nothing wrong with that, that's all part of any decision, there are a few perhaps hidden issues to consider, when looking at how fast one system is relative to the other, since you are looking at the mathematical square of the aperture ratio, small differences may matter.

The RASA 8 is a 400mm focal length optic with a 203mm front objective OUTER diameter. By the common definition that makes it a 400/203=1.97 or f/1.97 optic,rounded up Celestron call it an f/2 optic. Rightly so, as this typical definition goes, but if one is looking at a comparison between Reflectors and Refractors, and how much light is truly collected then it pays to consider how much light is lost/uncollected over the central obstruction. (Here I'm not considering any contrast loss, just how much light is not collected over the central obstruction and how that might equate something one could call the effective f-ratio.

The INNER diameter of the front objective (the central obstruction)is somewhat large at 93mm, so whilst the system is advertised as f/2, in light gathering terms it is a little slower, circa f/2.22. That is the number I would consider using in any comparison of exposure or light gather.

Another difference to consider is that the RASA 8 has a 32mm wide "Usable field" of which it defines 22mm as the Image circle, compared with 2 of the other telescopes you mentioned: The SW Espirit 100ED and 120ED, which have 40mm and 42mm image circles respectively. Of course this means you can use larger sensors, if you so desire, but also, as you've already noted you can use reducers to increase effective speed on an appropriate sensor.

So perhaps take that in to consideration when looking at how fast one system is relative to the other, for a given sensor.


Taking one of the examples you suggested, i.e: using the 0.65x reducer on the SW Espirit 100ED f/5.5 that would make it ~f/3.57 and reduce its image circle from 40mm to ~26mm. NOW comparing that with the RASA-8 at f/2.22 effective, that would make the RASA-8 about 2.6 times "faster" if using the entire 32mm "Usable Field" of the RASA-8. If using the 22mm Image circle of the RASA-8 things would be much closer, with the RASA-8 still being about 1.3 times faster, based on the reduction in area alone.


Best
JA

[Emuhead (Andrew)]

This is really helpful thankyou!

I also didnt know that by using reducers you effectively reduce the image circle as well. That would catch a lot of people out im sure.

I have pulled the trigger on the RASA 8, and so with my XT2 sensor (width 23.6mm) i think that wont be too bad at all.

And i def take your point on the central obstruction reducing the actual f/number.. in my case it will be reduced slightly more id say given the camera body will be larger than that again. Oh well.. ive dived in, and to be honest relieved for the moment that a decision has been made, lol.

I will def see how this goes and perhaps return to the alluring world of refractors at a later stage in this hobby.

[jahnpahwa (JP)]

Sorry to jump in here, but has anyone an opinion on the celestron edgeHD 8" with fastar at F2-ish? I guess this ends up being comparable to a RASA of the same size (though I assume RASA has different/superior flattening?) but has the added flexibility of being able to image at f10 and also be used visually?

I ask because I'm sort of in the same boat, thinking about an OTA to complement my 8" F5 newt... and figure that the 8"edgeHD gives me focal lengths either side of this.

(again, sorry to hijack!)

JP