So far / How to increase sub times?

[cadman342001 (Andy)]

Had some dark clear nights last week coinciding with the new moon so
I've had an opportunity to set up the SW Star Adventurer on the drive.

I should first of all point out that I have no knowledge of the night sky and previous attempts at photographing it have been 14mm views / stitches of the Milky Way.

Night 1.

Polar alignment being done by my Polemaster, even finding the SCP and sigma Octantis took me about 30 mins ! I had the rig pointing roughly south (compass) and set the lat/dec with the wedge adjustment (16 degrees here in Cairns). Ended up with Stellarium open (in night view), looking through bins, looking at Stellarium just trying to find octans, using the Snipping Tool in Win 10 to get a screen shot from the live view from the Polemaster software and plate solving with All Sky Plate Solver.
This usually showed the grid and therefore the location of the SCP.
I also turned on the marker for SCP in stellarium.

Once I found octans it was easy to use the polemaster software to do my PA although 1st night the SCP was in the midst of a load of dead tree branches of dead tree next door and then when I subsequently moved the rig I could only see half of octans above the roof line of the house ! enough to do the PA though as sig Oct was visible.
The joys of living so far north I suppose.


Nights 2 onwards

Enjoying practicing hopping my way around the night sky with the bins, checking against stellarium, looking for DSOs etc. in stellarium and then trying to locate them / nearby stars in the sky.

I began with an 80-200mm f/2.8 zoom, stopped down for sun stars and to allow ISO higher using the L-bracket and counterweight camera on ball head, seeing how long an exposure could be made. Since ditched the ball head and mounted directly on the "L" of the L-bracket for less weight, easier to counterbalance, less wind resistance and better tracking.

I'm getting nice pin point round stars at 2 minutes. If it's windy then probably 1 in 3 subs is trash due to star movement. I have a large aluminium Induro tripod (AT-413) and hand 3KG of weight off the hook in the centre but still getting movement.

I was doing 1 minute subs with longer FLs but getting no nebulosity showing up in the stacked files. Nice dark files with sharp stars but no space dust

I've attached the latest shot of Antares, with 180mm f/2.8 @ f/11, 149 lights, 52 darks, 20 bias frames. Will have to give this another go with better framing, not using the ball head is a bit more of a challenge framing wise.

Are 2 minute subs long enough to capture nebula ? I realise I'm probably getting some light pollution at home so dark skies will help. I know less focal length is one way to increase sub time but not practical for smaller DSOs, any other way to increase sub times ?

Oh, and I'm using DSS > StarTools > LR or PS CC

https://c1.staticflickr.com/5/4244/3...400167a5_b.jpgAntares 26 May STools by AndyMacDougallPhotography, on Flickr

Andy

[raymo]

Very nice, but why on earth did you use f/11? f/5.6 would eliminate most
lens aberrations, and grab more photons. What ISO did you use?
raymo

[cadman342001 (Andy)]

Yeah, probably a bit much ! ISO 3200. I'll try f/5.6 and ISO 800 next time ? or is 800 too low ? I read somewhere that say a Canon 5D mk3 has a sweet spot of ISO 1000/1100 so presumably it's similar for the 24mp D600 full frame Nikon ?

[raymo]

O.K. Andy, you need to experiment; use fewer lights while experimenting,
[say half as many] or it will take forever. Your noise situation looks quite good, so start by staying with ISO 3200, @ f/5.6,
then ISO 1600 " "
then ISO 800 " " .
Compare them all to see which gives the best result. Another night repeat
the exercise using f/4. Biases are normally unnecessary, as darks contain
bias information. You probably already know, but in case you don't, noise
will increase dramatically with increases in ambient temperature, so
experience will tell you when you you need to lower the ISO. The sweet spot
seems to vary between 800 and 1600 for Canon DSLRs, so again
experience will reveal your model's ideal ISO.
Hope this helps
raymo

[sil (Steve)]

Originally Posted by cadman342001

Yeah, probably a bit much ! ISO 3200. I'll try f/5.6 and ISO 800 next time ? or is 800 too low ? I read somewhere that say a Canon 5D mk3 has a sweet spot of ISO 1000/1100 so presumably it's similar for the 24mp D600 full frame Nikon ?

Too simplified way of thinking about it. When you "read somewhere" what optimum setting to use take it as a guide for setting to start TESTING from in order to find YOUR optimal settings for YOUR actual camera. No two "identical" will perform identically. You cant use generalisations or photographic tests and apply them to astrophotography. If you really want optimal settings its up to you to test, noting all factors, and find your own sweet spot.

[kens (Ken)]

You may need to use autoguiding to get longer subs @ 200mm.
Periodic error (PE) could be contributing to the in 3 subs you are losing.
You'll get about 2 pixels per minute of RA trails due to PE with that mount

[cadman342001 (Andy)]

Thanks everyone, all good.

Andy

[OzEclipse (Joe Cali)]

Andy,
There are 2 reasons for reducing ISO on a DSLR for astrophotography.
1. Reduce noise
2. Increase dynamic range

The sweet spot referred to in some articles is the point where reducing ISO doesn't increase dynamic range. It says more about the dynamic range performance of the camera than the high ISO noise performance. The lower the number the better the dynamic range.

A company called DxO also measures the noise performance of sensors. For a given low light level, which they don't specify but I've always assumed is higher than astro imaging levels. They quote an ISO for each camera sensor that corresponds to the ISO where the noise levels are -30dBrelative to signal. -30dB is a signal to noise ratio of 31. They call this the sports rating. So I think of the light level as being like I might find at a night time illuminated sports match not astro levels.

I find I can do astrophotography that I am happy with at -

maximum ISO = 2 x (DxO sports ISO rating).

But I only use this sort of high ISO for static tripod photography or for a moving target such as aurorae when necessary. I use 1x or 0.5 times where possible.

See http://joe-cali.com/eclipses/PAST/TS...ora/index.html

Others might not be happy with my ISO/noise tolerance. That's for each individual to decide. Of course you can go right down to the low ISO value and get better dynamic range but as Kens noted, you need to consider periodic error. I would add two other things -
1. polar alignment drift
2. an uncooled DSLR will suffer from thermal noise and it as not linear with time. The longer the exposure, the hotter the sensor gets, thermal noise quickly becomes a problem.

In Phil Hart's review of the Star Adventurer, http://philhart.com/content/star-adventurer-review, he measured the PE at 50". If you do long exposures that span a full PE cycle, you will need to use lenses 25mm and shorter to keep the PE wobble < 1 pixel without autoguiding.

In his review, Phil stated, "Once I moved up to 80mm focal length, I found I needed exposures of ~1 minute or less for pin-sharp stars and at 200mm I was using 30 second exposures. Two minute exposures at 200mm focal length were mostly trailed to varying degrees, but if you were just aiming for Facebook resolution then you'd still be fine. If you pixel peep and aim for perfection like I do then you'll need to stack lots of fairly short exposures for sharp images. But this 'track and stack' approach is well proven and is one I've used for a long time."


Your D600 camera has excellent noise characteristics.
DxO sports/low light ISO rating for D600 = ISO 2980
Highest ISO for highest dynamic range (sweet spot) ISO 200 Lower than this won't gain more dynamic range.

All astrophotography is a matter of compromises. The cheaper the gear, the more you need to compromise.

In your case,
The dynamic range ISO response of your camera results in improving dynamic range down to ISO 200 however polar alignment, large periodic error and thermal noise limit the length of exposures. Good high ISO noise performance allows you to experiment with high ISO's to shorten exposures to lengths that your gear can handle.

Lot's of good stuff on the web to self educate some aimed squarely at DLR's some more generally at noise :
References I used for info on the D600
http://dslr-astrophotography.com/iso...nikon-cameras/

https://www.dxomark.com/cameras/bran...taType=rankLln

Photographic Dynamic Range versus ISO Setting
http://www.photonstophotos.net/Chart...D,Nikon%20D800

Craig Stark of Nebulosity fame has an excellent youtube video of a talk https://www.youtube.com/watch?v=EO4QFb3ydNM

and a 5 part series of PDF's on signal to noise under the 2009 & 2010 sections of his articles pages.

Another excellent youtube talk on DSLR astrophotography by Tony Hallas is this one :
https://www.youtube.com/watch?v=PZoCJBLAYEs

good luck

Joe

[cadman342001 (Andy)]

Thanks for the detailed and informative reply Joe, appreciate you taking the time and thanks for the links, I'll follow them up.

So am I correct in thinking, keep the ISO as low as practically possible to preserve DR which is what I need to bring out faint detail when stretching the resulting stacked image without introducing artifacts ?

and that I'm aiming for a histogram in the individual shots with the peak halfway between left and the middle ?

Andy

[chromus (Phil)]

I think the major thing (as a newbie to AP myself) we first miss is the reason we stretch in post processing. We are looking to get on screen on the camera what we want as the end product and forget we are grabbing all the subs to make up for the "lack" of what we see on a single frame.

I tried ISO 3200 and couldnt even get them to stack reliably (750D) but the single frames looked pretty good.

I took the leap of faith based on similar advice to the above and went to IS0800 (sweet spot for my sensor) and got the reverse, the camera screen showed faint smudges, but the subs stacked very well and a quick simple stretch got me far better end results.

If you want to have a practice at stretching without the hassle of capturing the subs go here and download a package of lights/darks/etc

http://www.rawastrodata.com/

After looking at the lights I started to understand what I should see on the cam screen vs what I thought I should see.

[cadman342001 (Andy)]

Thanks Phil, sounds like exactly what I need ! What a great idea.

Andy

[OzEclipse (Joe Cali)]

Originally Posted by cadman342001

Thanks for the detailed and informative reply Joe, appreciate you taking the time and thanks for the links, I'll follow them up.

So am I correct in thinking, keep the ISO as low as practically possible to preserve DR which is what I need to bring out faint detail when stretching the resulting stacked image without introducing artifacts ?

and that I'm aiming for a histogram in the individual shots with the peak halfway between left and the middle ?

Andy

So am I correct in thinking, keep the ISO as low as practically possible to preserve DR which is what I need to bring out faint detail when stretching the resulting stacked image without introducing artifacts ?
Yes but don't be afraid to make it as high as necessary to keep subs short enough that PE, drift and thermal noise don't dominate and ruin the subs.

and that I'm aiming for a histogram in the individual shots with the peak halfway between left and the middle ?

There are different ways of dealing with subs and stacking. You can expose all subs as you describe above, then average all with a median or mean.
Ideally yes. Then the stacking process is mostly an averaging process. If subs are short on exposure you can group subs and add. Let's say the mount is only capable of untracked guiding for half the exposure you need. Group the subs into pairs then stack and add each of those pairs to get the intensity where you need it.

Then average the added files ie the results of all the files resulting from the added groups.

The videos and references I gave you are not specific to your situation, but are definitely worth looking at to get different ideas about how to organize capture and processing.

I also have to agree with Raymo that f11 was a big overkill on stopdown.

I have a Pentax K1 camera that is similar spec to yours. Full frame, but with smaller pixels (4.6µ), 36MPx sensor
DxO Sports ISO is 3280, Linear increase in dynamic range all the way to ISO100
My 50mm lens is f1.7. At f1.7 the coma is quite bad. But it is well controlled by f2.8 so I use it between f2.8 and up to f5.6

Look at :
http://joe-cali.com/astronomy/astrophotography-2/

The first image is with that lens at f5.6.
The 5th, 7th, and 8th images are with the lens at f2.8. Some coma visible in the far corners which I can live with for the extra light grasp that an f2.8 lens gives me.