Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
07-15-2010, 12:36 AM   #1
Forum Member




Join Date: Sep 2008
Location: Sydney
Posts: 91
Quick DFS/astrophotography question

I don't really see the big issue.. Well, I can see the issue, but IMO it's not that big?

I'm building a barn door mount, so essentially if I wanted say a 30 minute exposure, would I not just do 60 x 30 second exposures to avoid the mandatory DFS? Yes, I know you would then have the hassle of flicking the shutter on/off, but a cheap intervalometer can be easily built for around $45 - I have the required parts on the way now. Stacking can then be done as usual via RegiStax/DSS etc.

Without the barn door mount, exposures of anything more than 30 seconds would produce star trails anyway.

Soo.. What's the big problemo?

Oh - And is the DFS subtraction just as good as what RegiStax/DeepSkyStacker/Whatever would do anyways? If so, it's really just saving time, no?

Please tell me I've missed something!

07-15-2010, 01:10 AM   #2
Pentaxian
wildman's Avatar

Join Date: Jun 2008
Location: Flyover America
Posts: 4,313
QuoteOriginally posted by maxwolfie Quote
Please tell me I've missed something!
OK.

You are in the wrong forum?
Go back to Cloudy Nights.
07-15-2010, 01:25 AM   #3
Veteran Member




Join Date: Dec 2007
Posts: 923
QuoteOriginally posted by maxwolfie Quote
I don't really see the big issue.. Well, I can see the issue, but IMO it's not that big?

I'm building a barn door mount, so essentially if I wanted say a 30 minute exposure, would I not just do 60 x 30 second exposures to avoid the mandatory DFS? Yes, I know you would then have the hassle of flicking the shutter on/off, but a cheap intervalometer can be easily built for around $45 - I have the required parts on the way now. Stacking can then be done as usual via RegiStax/DSS etc.
That's true if the light is not too dim.

In really low-light applications, 30 seconds doesn't give you enough photons (light !) to build up image detail. Doing 60 x 30 seconds still doesn't build up enough details.

So you need more than 30 seconds, then you get forced DFS, and you can't capture anything that happens inbetween.
Example is a 1/2 moon-lit night landscape shot, with no other source of light...

QuoteOriginally posted by maxwolfie Quote

Without the barn door mount, exposures of anything more than 30 seconds would produce star trails anyway.

Soo.. What's the big problemo?
Really dedicated enthusiasts have clock-work or battery driven equatorial mounts which precisely counteract the rotation of the earth. A bit like what professional astronomers use.... I'm not into that kind of stuff, but I've seen them in operation before. The astronomy club in my local university has a monthly meeting, which I dropped in before. You see some serious gear there, all Nikons and Canons though...
07-15-2010, 02:11 AM   #4
Senior Member




Join Date: Sep 2009
Location: Beautiful Bavaria :-)
Posts: 123
A barn door mount allows only some minutes of exposures for unblurred pictures. Even with a high-precision motor mount you need permanent guidung control and correction for longer exposures.

The main question is: which type of celestial objects do you want to shoot and how is the environment where you shoot (light pollution is visible depending on your environment from 10 sec onwards!)?

For any kind of faint objects pictures, DFS is advisable, even for 10 sec shots!

07-15-2010, 04:15 AM   #5
Forum Member




Join Date: Sep 2008
Location: Sydney
Posts: 91
Original Poster
QuoteOriginally posted by wildman Quote
OK.

You are in the wrong forum?
Go back to Cloudy Nights.
Discussion about the K-x is pretty relevant IMO

QuoteOriginally posted by kittykat46 Quote
That's true if the light is not too dim.

In really low-light applications, 30 seconds doesn't give you enough photons (light !) to build up image detail. Doing 60 x 30 seconds still doesn't build up enough details.

So you need more than 30 seconds, then you get forced DFS, and you can't capture anything that happens inbetween.
Example is a 1/2 moon-lit night landscape shot, with no other source of light...



Really dedicated enthusiasts have clock-work or battery driven equatorial mounts which precisely counteract the rotation of the earth. A bit like what professional astronomers use.... I'm not into that kind of stuff, but I've seen them in operation before. The astronomy club in my local university has a monthly meeting, which I dropped in before. You see some serious gear there, all Nikons and Canons though...
Yeah, I guess that's the main issue. But really, if I'm going to be doing a dark frames regardless, I don't really see the issue - If anything it's a time saver? Sure, it doesn't give you control over how long the dark frame is (i.e. a 30 minute light frame exposure with a 30 min dark frame - a dark frame of equal length is surely overkill)

The result is still outputted as a RAW file isn't it? If you are shooting in RAW to begin with, of course.

QuoteOriginally posted by Vranx Quote
A barn door mount allows only some minutes of exposures for unblurred pictures. Even with a high-precision motor mount you need permanent guidung control and correction for longer exposures.

The main question is: which type of celestial objects do you want to shoot and how is the environment where you shoot (light pollution is visible depending on your environment from 10 sec onwards!)?

For any kind of faint objects pictures, DFS is advisable, even for 10 sec shots!
My intention is to build a double arm barn door mount - The claims are 15-20 minutes plus, but of course with longer focal length, my comparatively dodgy alignment and my very shoddy carpentry skills will no doubt shorten the maximum exposure time. I am hoping for 10 minute exposures at a reasonable focal length (i.e. 200mm). I'd be happy with that. I'd then stack a few 10 minute exposures in DSS and see how that goes

Is there an issue with stacking frames that have already had their dark frames subtracted? I.e. you take 3 x 1 minute exposures, obviously DFS kicks in after each exposure. Can I successfully stack the 3 exposures in DSS without losing SNR?

EDIT: I also don't understand why a 2 x TC will stop down the lens either. But I guess that's a bit of a separate issue
07-15-2010, 12:57 PM   #6
Senior Member




Join Date: Sep 2009
Location: Beautiful Bavaria :-)
Posts: 123
A teleconverter disperses the light entering through the front lens over twice the area in comparison to the rear element of the lens. The film/sensor size remains identical, so in terms of physics, 50% of the light are lost, because 50% of the photons have been guided to a place outside the film/sensor area.

OK, back to barn door mounts:

Sorry, I still don't get it:

- It is much better to stack 120 exposures of 30 seconds length: shorter exposure time reduces digital noise and minimizes risk of blur due to an unexpected gust of wind or similar issues. If you have artificial light sources nearer than ca. 50 kilometers, you'll end up with a totally overexposed 10 minute exposure (f 1.4 and 4 minutes at medium high ISO will result in green leaves of the nearby tree illuminated by starlight - try it!!!). Alternatively, light traces of bypassing planes are significant on a 10 minute exposure!

- I do not doubt your abilities as a carpenter or photograph, but deep sky objects or starfield pictures are the only astronomical objects that require long time exposures. In most cases guided exposures are the only way to give you satisfying results, even more if you consider the fact, that deep sky objects need large apertures and long(er) focal lengths to collect enough light and give enough magnification. So, which objects are you going to aim at? Hopefully, the barn door mount is exact enough to avoid blur by earth or image field rotation.

- The dark frame should resemble exactly the exposure conditions of the real pictrues taken: if you take 6x10 min, do at least 2 6x10 min dark frames (beginning and end of session. A much better way is to create a master dark frame by summing up some 30 or 50 df exposures with the identical exposure conditions, stack and average, now use the master df to denoise the stacked real picture by subtracting it.

I hope, this helps answering the question. Clear skies!!!!
07-15-2010, 02:59 PM   #7
Forum Member




Join Date: Sep 2008
Location: Sydney
Posts: 91
Original Poster
QuoteOriginally posted by Vranx Quote
A teleconverter disperses the light entering through the front lens over twice the area in comparison to the rear element of the lens. The film/sensor size remains identical, so in terms of physics, 50% of the light are lost, because 50% of the photons have been guided to a place outside the film/sensor area.
Thanks for that!

QuoteOriginally posted by Vranx Quote

OK, back to barn door mounts:

Sorry, I still don't get it:

- It is much better to stack 120 exposures of 30 seconds length: shorter exposure time reduces digital noise and minimizes risk of blur due to an unexpected gust of wind or similar issues. If you have artificial light sources nearer than ca. 50 kilometers, you'll end up with a totally overexposed 10 minute exposure (f 1.4 and 4 minutes at medium high ISO will result in green leaves of the nearby tree illuminated by starlight - try it!!!). Alternatively, light traces of bypassing planes are significant on a 10 minute exposure!
As kittykat46 was getting at (I think).. a 30 second exposure for DSO isn't really enough...Regardless of how many 30 second frames you actually do, as it then just becomes an average of poorly exposed frames (hence resulting in an average, poorly exposed frame with better SNR).. But yes I agree light pollution then becomes an issue in built up areas and shorter exposure times may be the only option.

I am thinking more along the lines of 5-10 minute exposures and the quality of the in camera DFS once this 5 - 10 minute exposure is complete. I am also wondering if it's possible to stack multiple 5-10 minute exposures in DSS or RegiStax (that have obviously already been processed by the DFS in-camera).

QuoteOriginally posted by Vranx Quote

- I do not doubt your abilities as a carpenter or photograph, but deep sky objects or starfield pictures are the only astronomical objects that require long time exposures. In most cases guided exposures are the only way to give you satisfying results, even more if you consider the fact, that deep sky objects need large apertures and long(er) focal lengths to collect enough light and give enough magnification. So, which objects are you going to aim at? Hopefully, the barn door mount is exact enough to avoid blur by earth or image field rotation.
I was hoping for some better shots of the Eta Carina nebula, as well as the nebulas found around Sagittarius/within the Milky Way (Trifid nebula etc.)


QuoteOriginally posted by Vranx Quote
- The dark frame should resemble exactly the exposure conditions of the real pictrues taken: if you take 6x10 min, do at least 2 6x10 min dark frames (beginning and end of session. A much better way is to create a master dark frame by summing up some 30 or 50 df exposures with the identical exposure conditions, stack and average, now use the master df to denoise the stacked real picture by subtracting it.

I hope, this helps answering the question. Clear skies!!!!
Wow, double the amount of exposure time just for a the master dark? I thought typically you would have a much longer light frame exposure rather than total dark exposure.. in the worst case, an equal amount (I.e. as the DFS functions on the camera)

So, in summary.. is DFS a critical failure for the K-x? It seems that yes, it does limit the ability to control your OWN DFS subtraction, but in the end, I can't see it causing that much (any?) grief. Saves me from going outside again to start the DFS myself.

Last edited by maxwolfie; 07-15-2010 at 03:05 PM.
07-15-2010, 04:11 PM   #8
Veteran Member
Ben_Edict's Avatar

Join Date: Jul 2007
Location: SouthWest "Regio"
Photos: Albums
Posts: 3,303
QuoteOriginally posted by Vranx Quote
A teleconverter disperses the light entering through the front lens over twice the area in comparison to the rear element of the lens. The film/sensor size remains identical, so in terms of physics, 50% of the light are lost, because 50% of the photons have been guided to a place outside the film/sensor area.
That depends on the magnification of the tc. Your doubled are is true for a 1.4x tc. (therefor the 1 f-stop "light loss"). With a 2x tc you'll "loose" 2 f-stops.

Also, the area-based explanation you give is misleading. Because it implies, that the lens-tc-combo could than be used on a larger format camera without loss of light.

The point is NOT, that the light falls "outside the sensor area"!

The easiest way to explain the "light loss" of a tc is: The lens focal length will be increased, but its open diameter (or entrance pupil) stays fixed. As the aperture is calculated by dividing the focal length through the open diameter, it is at once clear, that the aperture has to go down by 1 f-stop (1.4x tc) or 2 f-stops (2x tc).

Or: If the focal length is multiplied by 2, then the aperture setting must also be multiplied by 2.


QuoteOriginally posted by Vranx Quote
Sorry, I still don't get it:

- It is much better to stack 120 exposures of 30 seconds length: shorter exposure time reduces digital noise and minimizes risk of blur due to an unexpected gust of wind or similar issues. If you have artificial light sources nearer than ca. 50 kilometers, you'll end up with a totally overexposed 10 minute exposure (f 1.4 and 4 minutes at medium high ISO will result in green leaves of the nearby tree illuminated by starlight - try it!!!). Alternatively, light traces of bypassing planes are significant on a 10 minute exposure!
Yes and no! If the subject is faint, the signal to noise ratio will be too low with short exposures, to produce a useful image. No stacking can save such an underexposed image. Stacking can increase signal to noise ratio by averaging the noise out. But it cannot bring out details, that haven't been recorded in the single exposure.

Ben

07-15-2010, 06:03 PM   #9
Forum Member




Join Date: Sep 2008
Location: Sydney
Posts: 91
Original Poster
^ What Ben said I what I was trying to convey (poorly!)

So, those that are using a K-x/K7 or any other Pentax body with "forced" DFS - How are you finding it for exposures longer than 30 seconds - I.e. those with mounting systems similar to barn door mounts etc.

Is it a pain in the butt or does it actually save you some time/fiddling around? Or are you impartial to it?
07-15-2010, 11:22 PM - 1 Like   #10
Site Supporter
smigol's Avatar

Join Date: Sep 2008
Location: Menlo Park, CA
Photos: Albums
Posts: 391
Regarding the comments about DFS and building master darks, I have been doing this project with my K10D over some time.

Here are a few pointers that I've come to understand.

Before you burn your time in the field and get frustrated with poor results, understand your lenses and how they will reach focus at infinity. I only shoot with primes, helping to reduce the problems they cause. The best way I've found to get the perfect focus with using camera lenses is to find a bright star or planet (Venus or Jupiter are great). Then I put two rubber bands across the open end of the lens in an "X" pattern. Then when I take the test pictures, I will get diffraction spikes. These spikes should form crisp lines when critical focus has been reached. You may be surprised that "infinity" is actually not at the infinity mark on your lens.

When you start getting into the 5-10-15-20 minute exposure range, either by stacking and summing or individual exposures, you'll run into skyfog issues. This is when the background glow of the atmosphere starts to show up. This can happen even in dark sky areas, albeit slower than in light polluted areas. For the dark sky site I go to, I've found that individual exposures of 15 minutes at ISO 800 are as much as I can get on a very dry, clear night.

A major boost will be the addition of a light pollution filter. You can get a well-balanced 2 inch filter (I use the one from Baader Planetarium) with an adapter ring to put it on your 49 mm lenses. This will allow you to get some much longer exposures.

In addition to the sky fog, you'll start to notice gradients across the frame. This is because the background color of the sky does change. A sunset's colors are an example of these gradients. You'll do well to center your subject in the frame so that you can control the gradient in post processing.

When I started doing the photography with my K10D, I left the in camera noise reduction on because I thought that it saved a step for me later. What I found was that there were several times over the course of sessions where long exposures would be lost because the batteries would lose too much voltage in mid-NR. What I found was annoying was that the camera wouldn't "automagically" switch the power supply to the grip or body battery in mid-exposure. It seems to only do this switching between exposures.

When you do get into this, you may want to use the intervalometer functions or purchase one of these devices or run your camera tethered to a notebook. I've done both and am satisfied with running the notebook as it allows me to do autoguiding.

When doing this task, you will get much better improved performance from the camera and the batteries by letting it rest for about 2 minutes between shots. With two fully charged (albeit 2 year old) batteries, I have done 5+ hours of continuous exposures of lights and darks in 50F temperatures. This is more than enough time for me to wait for darkness, do drift polar alignment, wait for my subject to get at least 40 degrees up in the sky, and then shoot until my batteries in the laptop die and I run out of snacks. :-)

A note on autoguiding. When you are doing images at 200mm, an autoguider REALLY helps. This is getting ahead of where you are at the moment, as you would need a mount that would allow this tool. It will be much more forgiving if you start with a wide-field lens like a 50 mm.

I would highly suggest reviewing this book by Robert Reeves: "Wide Field Astrophotography: Exposing the Universe Starting with a Common Camera" It details the construction and operation of barn door mounts as well as what you can expect to get from basic cameras and setups. You can get the book from William-Bell Wide-Field Astrophotography, Exposing the Universe Starting with a Common Camera by Robert Reeves

Lastly, for an example of what I did with using in-camera noise reduction with long exposures and a D FA 100mm F2.8 Macro lens:

http://www.flickr.com/photos/smigol/3840664082/

Last edited by smigol; 07-15-2010 at 11:41 PM. Reason: Fixed a link
07-16-2010, 01:28 AM   #11
Senior Member




Join Date: Sep 2009
Location: Beautiful Bavaria :-)
Posts: 123
QuoteOriginally posted by Ben_Edict Quote
Also, the area-based explanation you give is misleading. Because it implies, that the lens-tc-combo could than be used on a larger format camera without loss of light.
The point is NOT, that the light falls "outside the sensor area"!
The easiest way to explain the "light loss" of a tc is: The lens focal length will be increased, but its open diameter (or entrance pupil) stays fixed.
Sorry, but I disagree: a teleconverter enlarges the aerial image at the focal point of the lens (a 2x TC doubles the image size). The focal length of the lens is not affected, but a lens with twice the focal length would give the same size of image field, that's why using a TC seems to double the focal length.
Believe it or not: the new aerial image DOES cover a larger image area, but of course the image is four times darker and therefore NOT projected without a loss of light, or better said: larger format yes (but not usable, because optical errors are as well magnified), but of course with drastic loss of light (and resolution!).
Now, we have reached the moment for a new thread, I believe...
07-16-2010, 05:17 AM   #12
Veteran Member
Ben_Edict's Avatar

Join Date: Jul 2007
Location: SouthWest "Regio"
Photos: Albums
Posts: 3,303
QuoteOriginally posted by Vranx Quote
Sorry, but I disagree: a teleconverter enlarges the aerial image at the focal point of the lens (a 2x TC doubles the image size). The focal length of the lens is not affected, but a lens with twice the focal length would give the same size of image field, that's why using a TC seems to double the focal length.
Believe it or not: the new aerial image DOES cover a larger image area, but of course the image is four times darker and therefore NOT projected without a loss of light, or better said: larger format yes (but not usable, because optical errors are as well magnified), but of course with drastic loss of light (and resolution!).
Now, we have reached the moment for a new thread, I believe...
Ofcourse the inverse square law applies to lenses and ofcourse the illuminated field will increase (until the lens designer prevented this mechanically). BUT your explanation is nevertheless incorrect:

It simply does NOT matter, whether any light will then fall outside the sensor area.

WHAT DOES MATTER is, that the subject's projection gets bigger through the tc and occupies a larger area ON the sensor! This means, the image is distributed over a larger part of the sensor and accordingly the light from the subject is distributed over the larger area and gets dimmer at each single point (the inverse aquare law).

A loss of resolution, by the way, is only a technical, but not a theoretical problem. The resolution is ONLY defined by the diameter of the entrance pupil and not by the focal length. So if a lens/tc combo looses resolution (compared to the lens itself), than the combination isn't a good match or the tc is not really good.

Really good tcs (at least some of the 1.4x modells) have such a good optical quality, that any resolution loss is negligible and more than compensated for by the increase of resolution on the sensor.

Ben
07-16-2010, 05:45 AM   #13
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
QuoteOriginally posted by Ben:
The resolution is ONLY defined by the diameter of the entrance pupil and not by the focal length...
Ben I think the diffraction limit of resolution is a function only of F-Number, ie. (1/(angle subtended by aperture)) times wavelength.

Last edited by newarts; 07-16-2010 at 07:01 AM.
07-16-2010, 06:02 AM   #14
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
QuoteOriginally posted by smigol:
.... Then I put two rubber bands across the open end of the lens in an "X" pattern. Then when I take the test pictures, I will get diffraction spikes. These spikes should form crisp lines when critical focus has been reached. ....
If I were to make such a setup with a filter ring or lens cap or lines scribed on a uv filter, should I use thick lines, like a rubber band, or thin lines like hairs?

BTW thanks for the informative post!

PS once infinity focus is determined for a lens/camera combination, must it be re-determined each time you use the lens?

Last edited by newarts; 07-16-2010 at 07:03 AM.
07-16-2010, 08:16 AM   #15
Site Supporter
smigol's Avatar

Join Date: Sep 2008
Location: Menlo Park, CA
Photos: Albums
Posts: 391
I use two thin rubber bands, as these are the most flexible. The ones I use come from bundles of onions at the grocery store, maybe 2 mm in diameter.



I do this exercise every time I use the lens, as temperature is different and I can never remember exactly the right position!
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
barn, camera, dfs, door, dslr, exposures, photography
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Quick question Raptorman Pentax DSLR Discussion 5 12-27-2009 11:03 AM
DFS question WillCarney Pentax DSLR Discussion 8 11-20-2009 06:03 PM
astrophotography equipment question sholtzma Pentax DSLR Discussion 0 05-18-2009 08:56 AM
Quick question ACImages Pentax SLR Lens Discussion 7 03-26-2009 11:56 AM
Quick Question Fl_Gulfer Post Your Photos! 2 04-16-2008 05:19 PM



All times are GMT -7. The time now is 12:01 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top