Will this work for a dodgy DSLR setup?

[Stonius]

I'm mostly doing visual at the moment.

Next time I'm at a dark sky site, is there any reason I couldn't just set up my stock Canon 5D Mkii on it's tripod doing interval exposures while Orion drifts through frame? I'd be using a Sigma Art 50mm f1.4, which equates to about 40 degrees of sky. With periodic realignments and sacrificing the frame edges, I should be able to get stackable subs, right?

Question is, what time? My research seems to suggest around 10 seconds is right, though this seems incredibly short, but what do I know? I figure I'll do a few test shots and zoom in to see how round the stars are.

I take it Darks, Lights and Flats are still required, even if shooting through a lens, as opposed to a scope?

Side question; I envisage eventually having an astrophotography rig (waaay down the track) that I set gathering data, while I do visual work. Is that a possibility, or does it require constant monitoring that would preclude any other activities?

Cheers

Markus

[silv]

Hey :)

astrophotography on a tripod?
here is what you need to know:
http://http://www.capturingthenight.com/astrophotography-and-the-600-rule/ (http://www.capturingthenight.com/astrophotography-and-the-600-rule/)


Darks flats etc: yes.
you do them because of the camera noise which has nothing to do with the lens used.

Speaking of which: a 50mm lens doesn't reach incredibly far so you will get a star field or wide field image. very beautiful!! and especially beautiful with your Sigma lens! (But you won't see the butterfly outline of the Orion nebula in the photos.)
Be sure to post your findings, okay?

waaaaaay down the track, you can do visual observing after having set up your imaging rig.
you'd just need an automatic camera shutter trigger so you don't have to click the remote yourself.

Have fun!!!!! :)
Annette

[pixelsaurus]

If you use Deep Sky Tracker, it will automatically align all your lights, darks and flats and produce a tiff for further processing.

http://deepskystacker.free.fr/english/index.html

[Stonius]

Thanks, Anette

Yes, I'm wanting to see if I can get Barnard's loop with stacking (I wouldn't expect to see it with a single 12 sec exposure). I'm just wondering how far one can take it without having to inve$t in too much gear :-)

Typical forum rules; I'll post it if it's good, and pretend it never happened if it's not!

:-)

Cheers and thanks,

Markus

[glend]

If you can include having a small laptop near your imaging rig then BYEOS provides many elements of Canon camera control, which go beyond what a simple interval metre can provide. You can download BYEOS to have a try.

[Camelopardalis]

Markus, go for it, it's a lot of fun!

The Sigma Art lenses are crackers. I have the 35mm and there is a lot of vignetting wide open, but it collects light insanely quickly :eyepop:

Take lots and lots of subs and you should be able to get something nice with stacking. The non-tracking tripod will do natural "dithering" and help with the noise, but I'd recommend taking some flats afterwards to help out with the vignetting.

I'm a bit of a pixel peeper, and up around the celestial equator wreaks all kind of havoc with star trails...with my 35mm, I found about 13 seconds was the max I was happy with. I'd expect to see star trails in shorter exposures with the 50mm. But, you might be surprised what you can get from ISO1600 or 3200 in 8-10 seconds ;)

[Stonius]

Thanks Dunk, That gives me a good frame of reference for my own experiments.

I thought (perhaps wrongly) that it is best to use the native ISO of the camera, in my case 800. Is that correct?

And thanks for the software suggestions everyone. I'll investigate them soon :-)

-Markus

[Camelopardalis]

Markus, it's a double-edged sword. On the one hand, we want our data to be as "pure" as possible, on the other hand we actually want to see some signal :lol:

800->1600 is only one stop, would probably be my pick. I've used 3200 on my 6D, which has a similar native ISO, and it works well. It'd be worth experimenting. Every camera, whether it be a point and shoot, DSLR or a dedicated astro cam has some degree of gain factor applied. It's just a question of finding the sweet spot.

The thing with increasing the ISO above native is that while you amplify the signal and some sources of noise, the read noise of the camera is constant (isn't amplified) so ends up effectively decreasing with respect to your (amplified) signal...

[Stonius]

Ah, that's good to know. Thanks for that. I will try the 1600 then as a jumping off point, I think.

:-)

[Camelopardalis]

Yeah, worth a try. The other thing to consider is that on warm summer nights the thermal noise will be amplified too...although my experience is that the full frame Canons tolerate the warm evenings a little better, for whatever reason.

[DarkKnight]

I've had a dabble with what you are proposing and find the '500 Rule' really is a rule of thumb. Your 50mm lens into 500 gives you a maximum exposure of 10 seconds and from my experience I'd go back to 5-7 seconds and take more subs.

I'd also suggest turning long exposure and high ISO noise reduction off 'in camera' and instead rely on your dedicated darks as it will eliminate the dark frames your camera automatically takes with each shot.

:thumbsup: for DSS, a good freebie to get you started with stacking.

Experiment with different ISO and aperture settings till you find the ones that work for you.

[sil]

Quite far. Take a look at my astrobin gallery (link in sig). Has a mix but its mostly Nikon DSLR on tripod. Nothing more fancy than that, except for the processing which is where the magic happens (and I'm redoing my workflow atm to improve what I've done with the same data). BTW I rarely bother with flats and darks and can get pretty much the same final output with just jpegs so ANY camera should be able to get similar shots. Not one of these shots looked anything like it in camera so forget trying to get it done there.

[silv]

and do me the favor and read the article I linked to in my post.

just in case you only want to read IIS answers to your question :P It's written by a member of IIS.

[CapturingTheNight]

Speak of the devil....;) Thanks for the link to my site Silv.


Hi Markus,
The 400/500/600 Rule as discussed in my blog post linked by Silv is a general guide to the maximum exposure time for a fixed tripod shot based on focal length and camera sensor size used, before the stars will begin to noticeably trail in the full image. Note- Full image. If you zoom in on any star image taken on a fixed tripod you will see that they will move after barely a second or so. So fixed tripod is always a trade off between gathering star light and star trailing. With a 50mm lens and FF camera the maximum I would ever go would be 10 seconds but I do a lot of just single image landscape astrophotography. As you are looking to stack the images to get detail in and around Orion, I would personally at least halve that to get tighter stars and take lots and lots of subs. I comfortably shoot my 5DmkII at 3200 and my 6D at 6400. As others have said I think you will be pleasantly surprised at how much detail you can get with say a single 5 second exposure, ISO 3200 and F/1.4 (or even stopped down a tad to F/1.8-F/2). The more subs you then take will only improve the final image detail. Just be careful to avoid terrestrial elements in your sub frames if you are looking to stack them. That is a level of processing hurt that you probably don't want at this stage.

One small step up from fixed tripod that I took early on before I bought a proper tracking mount for deep sky work- is I built myself a barndoor tracker for about $15 in materials and bits I already had lying around. I got up to 3 minute exposures with focal lengths up to 90mm with zero evident star movement. So if you are somewhat DIY handy it could be an option for you if you want an inexpensive next step after playing around with fixed tripod. These days there are off the shelf small trackers designed for DSLR cameras (Vixen Polarie, Skywatcher Star Adventurer, etc ).

Looking forward to what you can come up with.

Cheers

Greg

[Stonius]

Thanks for all the suggestions. I think I'll just have a go and see what happens based on that very handy 600(or 500) rule.

Im keen to try the barn door tracker, I found an arduino based one that smooths out the tangent errors, so I might have a go at that one, since I have an arduino lying around. With an extender I can get out to 400mm and with a barn door, I could do more than I'd hoped I'd be able to. Oh, no! Is this how it starts? Have I passed the the astrophotography event horizon without even knowing it? NOOOOOooooooooo............!

Exciting times!! :-)

Markus

[OzEclipse]

Hi Markus,

Greg's advice is very good and his wonderful results speak for themselves. I want to expand on some of his comments.

What ISO to use?

Greg is correct that you can use much higher ISO's than you are using. DxO Mark does noise tests on senors and comes up with a comparative ISO. Your 5Dii has a low light ISO value of 1815. I have two cameras, one is an APS C with a DxO mark rating of ISO1160, the other is ISO3200.

I can use the ISO1160 camera up to 6400 with noise reduction or stacking of 3-4 images

I can use the ISO3200 camera up to 12800 but I do get purple stars. At 6400, image quality is good. I would suggest that you should be able to use at least ISO3200 and probably ISO6400 with good results.

What is the maximum exposure time?

The 600 rule & its variants were derived for film a long time ago. It was derived by assuming that the viewer was looking at a 5x7 inch enlargement at arms length. Human eye can resolve about 1 arc min. 1 arc min at 800mm viewing distance is about 0.25mm on the print. Print is 5X enlarged so drift equates to about 50 microns drift on the film.

The sky moves at 15.04"/s and so(skipping some steps) it takes about 600/f seconds to cause that approximately 50 micron smear on the film at the celestial equator less as declination increases. Using the 600 rule, T(s) = 600/f will result in 50 micron streaks at the celestial equator. Fast film of yesteryear had grains that were 25-50 microns and so it gave the illusion of one star image on one grain.

But all of this calculation is pointless unless we also consider viewing distance and size.


Using the same 600 rule today, 45 microns results in 10 pixels of smear for a typical 4.5µm pixel DSLR viewing at 100%. But who views at 100% these days? My camera produces a 5000x7500 pixel image. I don't have a 7500 pixel screen. If you want to zoom into 100% to inspect details in images, you need to track and guide! But for wide field images, we typically want to enjoy the aesthetic of the whole image and so we can decide on a final display resolution and pick exposure time to suit.


A few years ago there was some discussion about this in these threads :

http://www.iceinspace.com.au/forum/showthread.php?t=105063
and this one
http://www.iceinspace.com.au/forum/showthread.php?t=102081

I derived a different formula for DSLR's that took into account, pixel size and screen viewing size. The original post is below and I've attached an Excel file that does the calculation for you. You can either do a specific calc in the top section or generate a table in the bottom. I've pre-populated both with info for your 5Dmkii.





Revised rule for DSLR's
These rules or formulae are based on a set of starting assumptions and have been kicking around for a long time.

In the old days of 35mm film where very fast films used for tripod astrophotography had very course grains and so a bigger movement could pass off as no drift. It was based on someone looking at a 5x7 inch enlargement at normal arm length reading distance perceiving trails as almost stationary. The eye can resolve about 1 arc minute so it can in fact resolve approximately 250 microns on a print which scales back to 50 microns on the film. Pixels on a modern DSLR are about 6µm. We mostly look at screen based images and images that are highly reduced from full size.

So I've derived a revised formula, modernized for DSLRs.

max exposure time(s) = [14*N*P] / [FL*cos(d)]


where
N..........Number of pixels of drift
to work this out load any full frame image from your camera and scale it to the size you want to view on screen. Look at the scale percentage. Number of pixels drift = 100 / percentage

P..........Pixel size (microns) Most DSLR's are between 3 and 8 microns. You can look it up in your manual or just split the difference and make it 6.5

FL.........Focal length(mm)

cos(d)...cos of the declination. Use the declination of the stars in the field that are closest to the celestial equator in the camera field

[Stonius]

Wow, Joe thanks for that incredibly informative post, and thanks for the calculator XLS.

I guess as you say, the end purpose of the shot must be taken into account. If you want it on your computer screen at 25%, then you can get away with a circle of confusion that's 4 x bigger than if you want to print it at full rez and stick it on your wall.

Makes sense! And I hadn't really considered the difference between the rate of movement of stars at the poles versus the celestial equator, but I see from the table it makes quite a difference!

Thanks

Markus

[sil]

Always remember the "rules" are really just a guideline for a starting point to take photos (subs) and will NOT themselves get you a clean bright colourful photo. I now work exclusively with camera on tripod, no telescopes or guiding. One true rule is not a single photo I take resembles the final photo I end up with (unless I'm shooting the moon). It all comes down to your processing workflow. You can get good photos with whatever gear you have if you understand how to process, which itself is a fluid thing and will change as you learn how to improve it. You dont NEED to take flats darks and bias frames (I rarely bother). Just start getting outside and photographing and processing any clear patch of sky.