Is Full Frame Better for Telescope Imaging?

[PembrokeSteve]

Hi,

This cloudy weather gets one browsing at cameras and drooling over what better cameras you might possibly buy in the future (this astro imaging hobby means = you are never happy with what you have!)
Anyway, I have a Canon 450D DSLR which I am very happy with, and it delivers me good image results through my Orion ED80 Refractor and Vixen VC200L VISAC telescopes.
MY QUERY:-
Would I notice much improvement (image quality wise)  if I were to pick up a secondhand "Full Frame" DSLR (ie) say a Canon EOS 5D? Would a Canon EOS 5D be much better than a Canon 450D?
Does a Full Frame camera only give a better result if doing "wide angle camera lens" shots of the milky way etc? or will a Full Frame sensor improve  image quality for subs through the Telescope?

Regards,
Steve

[vlaiv]

Both yes and no, or to put it in slightly different words: It depends

With full frame sensor there is more chance of: edge of the field aberrations - you need to make sure that scope is corrected for that size of field, use appropriate flattener (some flatteners are designed for aps-c size sensors and not full frame). Vignetting is also going to be more pronounced (depending on scope). So in general full frame sensor needs matching scope.

What does it bring in terms of performance? Well it actually depends on sensors in comparison and technologies, but in general - bigger pixel size - which is a good thing if you match it to a scope (lower resolution, but better SNR), bigger pixels also usually mean deeper full well capacity, so you will be able to do longer exposure without saturating pixels - again good for SNR, but you need to have good guiding/tracking in order to fully exploit it.

So if you plan to get a second hand full frame sensor, first look if your scopes will be able to fully illuminate it (with small vignetting), and also if optics is going to be well corrected for such large field. After that look into spec for particular sensor - being: pixel size, full well capacity, QE efficienty, read noise, dark current and compare to your current sensor. If you see improvements (better resolution match on your scopes, and better specs) - then go for it.

[wxsatuser]

A 5D would imo be a backward step, 2005 tech v the 2008 tech of your 450D.
The QE and read noise of a 5D are particularly bad compared to the 450D, although your Vixen should be able to cover a full frame.

 

[ollypenrice]

Sticking just to the question of chip size and ignoring chip technology it's pretty simple.

1) You need a telescope which can cover it. Most can't. The VC200L seems to have an image circle of 42mm which is not quite enough. (Quoted on the OPT website.) You really need about 45 to 46mm in my experience. (Tak claim 44mm for the Baby Q yet it doesn't even come close to covering full frame.) There is, no doubt, a difference between reality and some claims!

2) If your scope can cover it you get a wider field of view which is great when you want it but a waste when you don't (as on small individual targets.)

3) It means that on a given target you can use a longer focal length and so work in higher resolution. (With a full frame camera I can image a target in my TEC140 at 1000mm FL when, with a much smaller chip, I would have to use a much shorter FL with less resolution.

Olly

[wxsatuser]

3 minutes ago, ollypenrice said:

Sticking just to the question of chip size and ignoring chip technology it's pretty simple.

1) You need a telescope which can cover it. Most can't. The VC200L seems to have an image circle of 42mm which is not quite enough. (Quoted on the OPT website.) You really need about 45 to 46mm in my experience. (Tak claim 44mm for the Baby Q yet it doesn't even come close to covering full frame.) There is, no doubt, a difference between reality and some claims!

2) If your scope can cover it you get a wider field of view which is great when you want it but a waste when you don't (as on small individual targets.)

3) It means that on a given target you can use a longer focal length and so work in higher resolution. (With a full frame camera I can image a target in my TEC140 at 1000mm FL when, with a much smaller chip, I would have to use a much shorter FL with less resolution.

Olly

Thanks Olly, I was assuming to much.

Got caught with that one when I bought the Borg 67mm, it covers APS-C so they say, no it don't I say.

[ollypenrice]

59 minutes ago, wxsatuser said:

Thanks Olly, I was assuming to much.

Got caught with that one when I bought the Borg 67mm, it covers APS-C so they say, no it don't I say.

I think there are far too many false claims bandied about in astronomy circles - especially image circles!

Olly

[PembrokeSteve]

Many thanks for the advice given. Was later shocked to see how large the Canon EOS 5D actually is...it is like a built like a brick!
My Orion ED80 focuser would struggle with it's weight I think.
I will probably be better off sticking with my Canon 450D and getting it modified for about £90 quid.

Again many thanks,

Steve
 

[smr]

with the new sensor technology in crop sensor cameras like my 80D I don't think theres any need to go full frame, add to that dark current suppression and stacking you dont really need full frame imo, and the advantage of crop is that you get more reach. 

[pipnina]

14 hours ago, ollypenrice said:

The VC200L seems to have an image circle of 42mm which is not quite enough. (Quoted on the OPT website.) You really need about 45 to 46mm in my experience. (Tak claim 44mm for the Baby Q yet it doesn't even come close to covering full frame.) There is, no doubt, a difference between reality and some claims!

Could manufacturers be citing their 75% coverage? They might consider that "usable" (or exploitable) while the 100% coverage is somewhat smaller? Not uncommon I think for 75% ray to be twice the width of 100%.

[ollypenrice]

9 hours ago, pipnina said:

Could manufacturers be citing their 75% coverage? They might consider that "usable" (or exploitable) while the 100% coverage is somewhat smaller? Not uncommon I think for 75% ray to be twice the width of 100%.

If they do then they shouldn't! The whole point of an image circle, in my book, is to specify the chip diagonal it can cover. Takahashi claim the FSQ85 is 'suitable for 35mm sensosrs.' The actual size of the popluar CCD 'full frame' sensors varies with the source on which you find it but Atik, whose cameras I use, describe it as being 37x25mm, giving a diagonal of 45.22. As I said earlier, the FSQ85 doesn't get near to covering it. The claimed circle is 44mm so we can't fault Tak for this. We might have something to say about them referring to 35mm film in 2018 though!

Olly

[pipnina]

4 hours ago, ollypenrice said:

If they do then they shouldn't! The whole point of an image circle, in my book, is to specify the chip diagonal it can cover. Takahashi claim the FSQ85 is 'suitable for 35mm sensosrs.' The actual size of the popluar CCD 'full frame' sensors varies with the source on which you find it but Atik, whose cameras I use, describe it as being 37x25mm, giving a diagonal of 45.22. As I said earlier, the FSQ85 doesn't get near to covering it. The claimed circle is 44mm so we can't fault Tak for this. We might have something to say about them referring to 35mm film in 2018 though!

Olly

In all honesty, the lack of detailed specs for astro equipment sometimes astonishes me. If we're about to fork over £2000 for an OTA, you'd surely want to know just about every detail there is about the thing? You'd at least expect 100% and 75% ray diameter to be specified explicitly for an astrograph.