A new topic for all your astrophotography adventures! Smile Please post any cool images, tips and whatever else you deem appropriate.

Let me start right away with a short account of my own first steps in the realm of astrophotography:
So, I received a used but very decent camera, a Sony Cyber-shot DSC HX400V. It is by no means a camera for astrophotography. Nor is it a telescope, for that matter. But it comes very close. Razz
With a stunning zoom of 50x, or a focal length of 1200mm, I quickly discovered I could zoom in on the moon nicely. That was when I took my first astrophotograph. For a few months, I've been taking shots of the moon in this phase and that - I posted one of the best images here.
I did some research and eventually stumbled upon the blog BudgetAstro.net. There I learned about stacking. And finally, two days ago, I decided to try it out - I documented my attempt here. Now, I don't have much equipment - just a tripod, a few filters and the camera, not even a remote shutter control - so I was dealing with lots of motion blur, and the results were not quite satisfying. Cemetech admin Alex suggested I use the self timer to avoid motion blur when triggering the shutter. I did some research yesterday and found out that my camera can do better - I can use my smartphone as a remote control. Very Happy

With this in mind, I had another attempt yesterday and last night. And this time I went further than just the moon. Wink
During the day I took a few shots of the sun. Unfortunately, it was cloudy, so the sun was partially obstructed. Then, my filter (I used two and three relatively rotated polarization filters to create a strong darkening effect) seems to absorb longer wavelengths better - the image turned purple. There was still some lens flaring, and something was causing a blur.



The stuff on the bottom of the sun are clouds. I tried to pick a moment when there were none obstructing the sun, but it seems I missed it by a few seconds. Sad
I enhanced the image slightly to get this:



Do you see that speck of dust on the sun? This is (drumroll) Region 2670, a sunspot! Very Happy


Later, tonight, I attempted a few moon pictures and gave stacking another try. This time, I used software: AutoStakkert! does the job pretty nicely. Most places recommend Deep Sky Stacker, but their page explicitly says that it is not suitable for planets, while AutoStakkert! was developed exactly for this job. I have to say, the result turned out much better than last time, although you can see some artifacts on the moon's edge (zoom in at about 1 or 2 o'clock, for example):



Unprocessed frame right out of the camera for noise comparison:



And after I was done with the moon images, I went on a tour to find Andromeda, also called M31, as it is the 31st entry in the famous Messier catalog. I did not manage to properly photograph M31, you can barely resolve it bigger than a speck of dust. But I was not using the camera's full potential - much rather I had navigation troubles. I could not target the area properly, thus I had to keep zoomed out not to loose my location. Here, the Android app Star Map, the desktop program Stellarium and the web app astrometry.net proved incredibly useful - the Star Map app allows you to see star names and constellations in an augumented reality view, Stellarium is a very detailed star map for desktops, and astrometry.net allows you to upload pictures, recognizes the star pattern there, and spits out detailed location data and what is shown on the picture.
Anyway I ended up taking a few images of Cassiopeia and some extremely noisy pictures for my own orientation (I could like quarter the exposure time by ramping up ISO, so I did that to speed up "navigation"). The blur seen in the sun images persisted. I believe it is due to the way I focused - while I used the autofocus for the moon, I used manual focus for the stars and sun and I turned it all the way to the maximum distance, which probably goes a bit beyond infinity. I will try to focus a distant light or something next time instead.
In the end I compiled a small "guide" or "map" for finding the Andromeda Galaxy on the sky, as a replacement for no proper pictures. All you need to be able to do in advance is recognizing the constellation of Cassiopeia. Smile



The green lines and labels name the constellations you're looking for, the yellow lines and labels are imaginary lines that help (at least me personally) to locate the final target, and the red ellipse shows where M31 indeed is.
When you found Cassiopeia, you take the sharper of the two "arrows" and see where it points. In that direction, you will find a perpendicular line of three bright stars, and on it, an "L" shape of medium-bright stars. Right in the middle below the L, you'll see a faint, diffuse glow - that is the Andromeda Galaxy.

I hope this post was not too long and contained enough information for anyone to replicate what I did, and maybe we will eventually get to take photos of some nebula or galaxy together! Smile
Those moon photos are phenomenal. I meant to get out and capture the sun spot today but I got occupied with other things. So maybe on Wednesday. If you really want to get into photography planning, check out an app called PhotoPills. It's a $10USD app, and it's the best I've ever spent. It doesn't tell you star locations (nor planets) but you can use it to calculate tons of stuff, such as: Subject Distance, the max time to photograph stars without them trailing, the longest shutter you can have to get star trails of varying lengths, field of view, depth of field, etc etc. I mostly use it to track the Milky Way and plan photo shoots. I'll plunk my marker at a destination and use Google Images on my computer to visualize what I'm seeing in the app. There's an AR mode as well. It'll even tell you when the golden and blue hour is each day and stuff. It's a really handy app.

Secondly, a free app is Google Earth for the Desktop. It's a little funky for me but it includes a simulation of the night sky. So you can go anywhere in the world, select ground view (not street view) and then move the time forward until it's night or whatever time you want. I use Google Earth in conjunction with Google SketchUp to create 3D representations of real world landmarks. For example, I downloaded trees and put them in a field so I can virtually walk around and see where a good angle would be. I downloaded and slightly modified a lighthouse so I could see when the Milky Way would be directly behind it. I say that it's funky because the sky view is always 3 hours off. I might be simulating 9pm but the sky is still daylight. So I need to fast forward 3 hours to see the correct amount of light but now the Milky Way is in the wrong part of the sky. I don't know how to fix it. I've gotten no replies on the Google Earth Help Group, changing time zones doesn't do anything nor does uninstalling and reinstalling program.

Both these apps serve different purposes for Astrophotography but are equally helpful. I wish Google Earth had a mode where you can simulate a camera lens so that way I can see things as if I were looking through my 14mm. Haha.

If you had to get one, get Google Earth. It's a powerful tool for being free. You can download custom layers that show the levels of light pollution over an area as well as a bunch of other info. Then once you start traveling and planning your moon, star and/or, milky way shots a few months in advance get PhotoPills.
I was on vacations for the last almost-three weeks and happened to get another nice picture of the moon. I did attempt getting something of the milky way, but both attempts were complete failures (One with moderately high ISO and one with max ISO intended for stacking. The first one did not show anything and the second one was a uniform grey, and stacking did not save it.).

Anyway, back to the moon. I found that the half moon, as I happened to photograph it here, is practically the most ideal phase for images. While it is not as mysterious as the first image I posted in this thread, it is of much higher quality. Despite a significantly longer exposure time compared to full moon photography, the half moon turns out better, because the sun shines from the side and reveals the various surface structures that cannot be seen when they don't throw long shadows. Also, this time I gave more attention to the white balance of the image. It seems more... uh, natural, perhaps. And finally, this time the stacking seems to have gone perfectly - I couldn't spot a fault left by the software. I guess I am getting the hang of it. Very Happy

That's incredible! Nice and sharp compared to your earlier photos! Was that because of the shadows or because of the stacking? Either way, good work!

Nik wrote:
I did attempt getting something of the milky way, but both attempts were complete failures (One with moderately high ISO and one with max ISO intended for stacking. The first one did not show anything and the second one was a uniform grey, and stacking did not save it.).


What were your shutter and aperture? Stacking won't make data from nothing. Based on the "uniform gray" it sounds like your shutter was not nearly long enough.
I finally managed to get a shot of Andromeda!

The weather was cloudy (if not rainy or stormy) for the past month, so star gazing was not possible. Just in time for this weekend, however, the sky cleared up and I decided to have another attempt at catching the galaxy. Friday's session didn't go very well, but I found another method of precisely pointing the camera at something. My tripod has three extensible legs, and it turns out, I have much more control over the camera's orientation if I rotate it by adjusting the leg's length rather than how you normally rotate cameras on a tripod.
This eventually allowed me to point the camera precisely enough to capture Andromeda yesterday:


(see the blur in the middle - that's it at maximum resolution.)

I had to fight a lot of things. I put the tripod out on my window sill and pointed it roughly in the direction Andromeda should be, with a really wide field of view. I then gradually zoomed in and corrected my orientation using the tripod legs and locating Andromeda via the path described a few posts above (it was not really visible with the bare eyes, although I think I caught a glimpse once or twice because I knew where precisely to look at). At first I had some motion blur in my images. It was inconsistent, and thus not star trails. I managed to solve it by (wait for it) not leaning against the wall. Razz
The next issue, which I still have not properly solved, is that I cannot really focus my camera on infinity. Even this image is blurry. I managed to slightly lower the effects by closing my aperture, but this means I need longer exposure times and/or higher ISO numbers...
And then I found out that with enough zoom (this image has a field of view of 1.068° according to astrometry.net) I started seeing star trails even with an exposure of less then 10 seconds.
I eventually settled on an exposure time of 4 seconds, but the star trail is clearly visible here.

With an ISO of 2000, an aperture of f/6.3 and an exposure of 4 seconds, the image is still well underexposed. I did not dare to ramp up my ISO even more, as at some point, it just makes the image grey from noise (yes, it's the noise, Alex... the exposure was 30 seconds in that image I mentioned in my last post.). One thing I have to mention in my defense is that the images did not look this underexposed on my camera at night. I saw it could use some more light, but the details were not nearly as invisible as they are now, on my monitor, at day.

I took a few - as astrophotographers call them - "lights", of which I chose 16 to stack. Those are the frames with the actual image. In addition to that, there were five "darks": frames with the same settings as the actual image, but with the cap on, preventing any light from reaching the sensor array. These contain the data of various defects on your sensor, e.g. slightly reddish thermal things and whatnot. I stacked each of those by hand this time, and I think it even worked decently well. Then, I subtracted the dark stack from the light stack. Finally I adjusted the histogram, contrast and brightness to make the galaxy just stand out more and got this:



You can still see some red noise while the green and blue one disappeared when I subtracted the darks. How did this even happen? And is there a similar trick to remove the red noise?

And then I noticed how truly underexposed the image was... I opened Stellarium and navigated to the same field of view and time I was looking at yesterday evening.


(I rotated and rescaled the image slightly to align with mine.)

Turns out, what you can see on my photograph is not the galaxy. It is not even the nucleus. It is just the very innermost part of the nucleus. So, to actually see the galaxy, you do not have to zoom in more, as previously assumed, you need to get tracking equipment and expose longer. Sad


So I guess that settles the Andromeda thing for now. I may try to do this again with a bigger ISO, a wider aperture - if I figure out the focus problem - and bigger field of view to reduce star trails, but I will never manage to get a decently exposed image with my 4, or even 10 second limit.

What I still have some hope for, however, is another thing. Around spring, the Orion constellation is high up in the evening skies for me (currently it is only at something like 4 AM. I am not that dedicated. Razz). Aside from a nice constellation to take pictures of, there is a wide variety of easy to find nebulae, and the first and foremost of them bright enough to see with the naked eye, even if it only looks like a small diffuse star. I hope this is bright enough to catch without further tracking equipment with at least some detail. Until then, I doubt I will get to take images of anything but the moon, unless anyone of you guys has a suggestion.
I'd argue that you're probably reaching the limitations of the Sony Cyber-shot DSC HX400V you have. The sensor in that camera is pretty small, it seems to be 6.17x4.55mm, which in imperial is 0.24x0.18in. A Canon 750D (aka T6i) is what's known as a "Crop Body" and has a sensor size of 22.3x14.9mm or 0.88x0.49in. On my camera, a Canon 6D, which is known as a "Full Frame" has the sensor size of 35.8x23.9mm or 1.41x0.94in.

Most Crop Body cameras have the same sensor size, Nikon makes their crop body sensors a hair bigger. Canon Cameras have a "crop factor" of 1.6x to that of a full frame camera while Nikons have a 1.5x crop factor.

That's a huge difference and in astrophotography light is everything.

What does the sensor have to do with light? Well...

Quote:
And then I found out that with enough zoom (this image has a field of view of 1.068° according to astrometry.net) I started seeing star trails even with an exposure of less then 10 seconds.


That's The 500 Rule for you. The 500 Rule is a pretty accepted mathematical rule for astrophotography. There's also The 600 Rule. Basically, for a DSLR take the 500 and divide it by the focal length. If you have a crop body camera, you have to divide that by the bodies crop factor, so:

If you use a 300mm lens, 500/300=1.67s and divide that by 1.6 = 1.04s. You've got a 0.63 second difference between full frame and crop body. So, let's find out the "Crop Factor" of your Cyber-Shot:

The surface area of the Full Frame camera is: 855.62
The surface area of the Canon Crop body is: 332.27
The surface area of the Sony Cyber-Shot is: 28.0735

When I divide 855.62 by 332.27 I get 2.5750744876, which rounded to the nearest tenth is 2.6 Since we need to find the difference in area between the two, we can subtract the crop body sensor from the full frame. If we convert this to a percentage, and 100% is the Crop Body camera, then we can reduce that the Full Frame camera has 160% more pixels then the Crop body (260%-100%=160%), and in decimal form that's 1.6.

For the Cyber-Shot, 30.4778527793 or 30.5. 3,050% (30.5*100). Subtract 100% (which is the size of the CyberShot sensor) and we have 2,950 or 29.5x crop factor. By that math, you'd be able to have a shutter speed of 0.05649717514s, or 0.06s, or roughly 1/20th of a second. Which, realistically, isn't feasible.

So, yes. In a sense a larger sensor will capture more light by allowing for slower shutter speeds.

Don't go running out buying a DSLR, instead buy a motorized telescope! It'll be about the same price and you can use it to photograph deep space objects, aka DSOs. Since it's motorized it'll track the stars, allowing you to take exceptionally long photographs at an insane focal length. Celestron makes some pretty nice motorized telescopes, check out their NexStar line. You'll still need a camera though; A lot of people still buy DSLRs to go with it but you can probably go buy or use an old web cam. Which has perks and disadvantages:

Pro:
  • Fast frame rate, upwards of 60fps.
  • Cheap
  • "Portable"

Cons
  • Small image, web cams are usually 2MP.
  • Third party software support may be limited
  • Will likely be a video you'll need to export as individual frames.
  • Will need a computer to start capturing


Where as with a DSLR attached to a telescope:
Pros:
  • Much larger photos
  • Can shoot in RAW

Cons:
  • Bulky
  • Expensive
  • Heavy


If you couple a laptop with the web cam (which you'd likely need), the whole set up is a little less portable. You're limited to the battery on the laptop rather than the batteries for the DSLR. But with the DSLR, you may need to add some counter weight to the telescope.

If you're serious, start with the web cam. It'll be cheaper and you can get a technique down while saving up for a DSLR.
That was a very long post that basically boils down to "Alex proclaims that his camera is superior"
allynfolksjr wrote:
That was a very long post that basically boils down to "Alex proclaims that his camera is superior"


Far from it actually! When shooting DSOs any camera is going to have problems related to the the angular rotation of the sky, per The 500 Rule. So no matter what camera is used the photographer is going to need a motorized mount. A DSLR may be better than a point in shoot in terms of light capture but you won't be capturing any meaningful photos at either 1/20th or 1s.

Many, many astro photos are captured through accumulation of hours of photography, sometimes over a few nights. Upwards of 60 photos with shutter speeds around 5 minutes. If you want incredible color photos, you need to repeat that for each color filter: Red, Yellow, and Blue for a total of 180 photos in this example for about 15 hours of photographs. If the DSO you're photographing rises at 3a and first light happens around 6am, that's 3 hours of photography per night. So, you'd need to go out for 5 nights to capture all that.

There's no way I'd be able to get a significantly better shot of Andromeda than Nik without the aid of a motorized mount. Even then he's got more practice with technique and application of those techniques. I don't even know what's involved in subtracting the dark frames from the light frames. That said, if I were to go out tonight I'd bet my photo wouldn't be any better than his.

Even then, a lot of folks actually use web cams for their cameras because they're easier to use. You can get a live view on a laptop screen and there's specialized software available that'll take the video and turn it into individual frames. You don't really need to shoot in RAW when pointing at DSOs because the light data is in the form of shorter photographs; If the photo is too bright then remove a dozen or so frames. Plus, as a downside with a DSLR, if you need to change camera settings you have a slight chance of messing with the alignment. In that regard, a web cam is much more straight forward as all the settings are controlled via the software on the laptop. The only real advantage to a DSLR is image size but that can be achieved via webcam by creating a mosaic of the DSO, you'll need a pretty high power telescope and tons more time because now you gotta capture each tile of the mosaic in all the color filters... but it's doable.

60 photos per channel is probably the extreme end of things but it's not unheard of. .
That's complex... My version of astrophotography is a lot "worse": put a camera behind the end of a telescope, point and shoot! Of course, me and my $15 telescope (~30x magnification, iirc) and a phone camera aren't going very far... Smile
I just felt like bumping this topic. I was recently out for a night taking pictures with some friends, and for calibration purposes took an image of Venus, as a really bright object in the sky. The original purpose was to try and test a different focusing method. But guess what I found out when I later looked at it?



I wouldn't have expected that I could actually resolve Venus, the camera is the same as the one I used for my past imaged. And you can really see that it is at its optical limit, with the chromatic aberration. Nevertheless I found the fact exciting, and it gave me a better idea of just how small or large Venus can be. Not to mention that I am getting the hang of properly manually focusing my camera to infinity in almost complete darkness, which is harder than it sounds.
Nice work! What lens and settings is that?
Lens is the built-in (and not changeable) one from my Sony Cyber-Shot HX400V. That camera is quite outdated these days, though, I wouldn't recommend buying it. Wink
Settings are:
Max. focal length, i.e. 1200mm according to the labeling, which I do not quite believe to be true, though. 215mm according to file metadata.
1/250 exposure time
F/6.3
ISO 320

All according to file metadata. This is in no way fine tuned, though. The picture is no good quality wise either. It was really just a snapshot towards a bright object to test just how much out of focus it was (not much, it seems). And the only reason I am shared it here is because I was rather surprised to actually see the disk of Venus on it.

I don't think I will be doing more Venus shots, it is just too small to capture without a telescope. FYI, the image I posted is heavily cropped. Let me get this into perspective: The vertical resolution of my camera's images is 3456 px. The moon in my older images is about 1/2 or 1/3 of that in height. The disk of Venus in this one is just under 40 px in height, which is smaller than 1/90 of the vertical size. And, I mean, it's not even 40 pixels large, and the chromatic aberration at this level of magnification is insane (even though that's actually just two or three pixels wide). It's really pushing the limits of the camera as is, and it probably won't get any better. If you want to take pictures of planets, you should get yourself a decent telescope which may, in fact, be cheaper than my HX400V was when it was bought. You could probably find something acceptable for $300-400 and use it with any old DSLR you may have laying about.
FWIW, it's likely a 1200mm FULL FRAME equivalent. The camera has a 4.3-215mm lens, and I can tell you that a 4.3mm lens on a full frame camera is super, super wide; I have an 8mm and that's has like 178º field of view.

With it's 50x optical zoom, 4.3*50 is 215. If the packaging since 1200mm and divide 1200/50 we get a lens with a 24mm-1200mm equivalent to a full frame lens.

It's a great picture Smile
As you may or may not have heard, currently comet NEOWISE F3 has become significantly brighter than expected, to the point of being reasonably well visible to the naked eye. It can currently be seen early in the mornings, after midnight. It is expected to be at peak brightness in about two weeks from now, around June 20th - 25th. At this time it will be up even earlier, so that we may be able to see it before midnight. As a bonus, the moon will be new and dark.
That said, keep in mind that comets are very unpredictable and there is no guarantee that it won't suddenly break apart or cease activity and become too dark to see sooner than anticipated. Provided that does not happen, I am hoping to go out for a night or two around peak brightness time and try to get some nice images.

Anyone else planning to watch it? In case you are interested, here is where you will be able to find the comet over time:


5 second exposure, f/5 aperture, ISO 500, 960 mm focal length. All in all I have taken over 60 frames this morning between 3:30 and 4:00 AM, with a total exposure time of 5-6 min. Of these I used 24 here, totaling 2 minutes worth of exposure. Stacked them with Deep Sky Stacker, which would not recognize the comet in the pictures so I had to mark it manually in every single one of them. Some light post processing with Photoshop (cropped, adjusted brightness levels, reduced the noise). Overall I am satisfied with the result, even though it still is grainy and slightly blurry. I will try again in a few days if the comet remains bright enough.
Very nice! I'm hoping to photograph NEOWISE soon but lack the early morning commitment and a clear shot of the sky Sad

I was racking my head why you couldn't just use the stars to stack the images then realized that the comet would not necessarily be traveling in the same direction of the stars.

Very nice shot and I appreciate the description of the steps on how you got the photo.
Alex wrote:
the comet would not necessarily be traveling in the same direction of the stars
Actually it is much more stupid. Over the course of half an hour the comet travels a negligible distance against the background stars , if you mean that. It would probably not cause any visible star trails at this focal length. The actual reason is that I don't really understand the inner workings of DSS. It would keep picking up gibberish stars given the bright sky and heavy noise of my raw frames (well, jpeg, but unprocessed). I neither had the time nor the will to fix the star detection and my understanding of the software is not enough to work around that, so I was getting an output even less crisp than it is right now (misalignment? I don't know. It was not horribly misaligned, but visibly blurred.)

I basically told it to ignore all stars, manually overrode the comet detection algorithm and pointed it at where exactly to look for the comet.
As far as I understand, had it moved faster against the stars, this approach would have produced star trails. Not the ones visible right now caused by the Earths rotation, but star trails caused by "tracking" the actual motion of the comet. Given that the comet is exceptionally bright, though, my exposures happened in a short frame of time and this technique worked out best for me. If I take any pictures, say, tomorrow, I probably won't be able to stack them with the ones from yesterday using the same approach, though.

YMMV, and if you decide to attempt a similar thing with DSS keep in mind that it definitely involves a fair bit of experimentation with it's various image alignment and stacking algorithms. Here I used the bilinear alignment and median stacking methods. Perhaps the ability to judge which to use comes with experience, I just played around with all kinds of parameters instead.
Also, to save you the trouble just in case, sanitize your frames manually first (discard shots with blur, stray planes, etc.). DSS can judge picture quality and discard the worst few percent, but just as the star and comet detection did, that returned weird results for me. It is made for deep sky images, after all. I am basically abusing it by feeding it early sunrise comet tail shots.

Alex wrote:
Very nice shot and I appreciate the description of the steps on how you got the photo.
Thanks, and please let me know if/when you attempt to get a shot of the comet. I'd love to exchange experiences. This is a unique situation for me, because I basically have the first opportunity ever to take a proper shot of an astronomical object that is not the moon, and I feel like it is not beyond the technical limitations of my hardware. Very Happy

Also to anyone else stumbling by here -this comet is the brightest in years and probably will be the brightest for years to come. Don't miss the chance to watch it! Smile
I've gone out twice now and gotten photos! Super cool. Last night I took about 800 photos. A few separate time lapses with the 6D and some close up shots of the comet with the T2i that I plan to stack and stuff. I somehow managed to take -23 10s Dark Flat photos though. The time lapses are coming along great. They're not amazing but that's expected.

I do plan to use some of the T2i photos in the time-lapse but they were shot in infrared because I packed the wrong camera (thankfully the battery was charged) so... yeah.

Here is an unedited photo from my iPhone, you can see Saturn in the middle of the picture, then Jupiter to the right, and also some really bright big rock of some sort.



Unfortunately I haven't been able to see comet neowise yet but I really want to. Maybe I have to get out of town or just find an open field or something? I'm in Oklahoma and have been looking to the northwest about an hour after sunset, and I haven't been able to see anything.
I recommend an astronomy app for your phone. I use Night Sky on iOS (unsure of an Android alternative) and was able to find the comet super easily after that. I also had a very hard time seeing it with my eyes while others could see it easily, however.

Additionally, the comet is VERY faint compared to other stars. I can still see stars just fine but the comet is difficult.

Below, are two photos I took a few days apart. Hour wise, these photos have a 45 minute gap (11:15p vs 10:30p), and the comet doesn't move that fast, so expect to look more progressively West as time goes on. Second photo is pretty much due NW. Both photos from the same parking lot, but maybe 50ft between the spots and different lenses)

First night 07/16 @ 11:15pm (85mm):


Second night, 07/20 @ 10:30pm (35mm):
  
Register to Join the Conversation
Have your own thoughts to add to this or any other topic? Want to ask a question, offer a suggestion, share your own programs and projects, upload a file to the file archives, get help with calculator and computer programming, or simply chat with like-minded coders and tech and calculator enthusiasts via the site-wide AJAX SAX widget? Registration for a free Cemetech account only takes a minute.

» Go to Registration page
Page 1 of 2
» All times are UTC - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement