I have been a semi pro photographer for over 35 years, so I know my way around photography and gear. I just recently decided to try and live my life long dream of doing some astrophotography. I have done a few wide angle shots of the Milky Way and Comet Neowise, but really want to get something of Andromeda Galaxy (m31) and eventually move on to other deep sky subjects. I am using good gear and have read as much as I can on the subject, but I am still failing miserably! The image below is the best I could do.
For M31 I am using the Sky Watcher Star Adventurer, with the balance weight, a Nikon D810 and either a Tamron 70-200 f2.8 or a Nikon 300 f4 PF. I am doing my best at aligning the Star Adventurer to Polaris and have tried various exposure times and ISO settings. Last night I used the 300 f4 PF lens at f4 and shot 15 frames at 2 minutes and ISO 2500. The files are all very bright and washed out, but I loaded them all onto Deep Sky Stacker with the "dark" "flat" and "bias" files I created earlier. The stack in DSS is still really bright white and washed out, but when I load the saved tiff into Photoshop and process it, I can get something that looks like M31 and it's sister galaxy. However, what I am getting is unacceptable and far worse than anything else I can find online, and even much worse than I have seen from people with lesser gear and no clock drive. I am at a loss as to what I am doing wrong and am really hoping that the people here can help me with this.
I live on a 2 acre property about 12 km outside of the city of Victoria (population about 400,000) so there is still a reasonable amount of light pollution here. Could that be my problem? I have seen pretty good samples online where people "say" they shot M31 from within a city and light pollution and I would be very happy to get something as good as they display. Do I need more frames or slower/faster exposures? What can I try next?
Any and all advice/help is greatly appreciated. :)
Cheers
Tim
From what I've seen, for a dslr shot of Andromeda, it's all about time on target.. like a few hours a night over the course of a month. Compile in batches then compile batches of the batches.
Perhaps even record in video, and compile the video in Registax? or even PIPP and AutoStakkert.
What you have is far better then what I could get using a d5500 and a 70-300mm kit lens. 1s exposures, and about a minutes worth of exposures. No tracking. I could clean it up more but I kind of like this vintage look in a way. I have yet to get a good clear night since getting my Z6.
Do you know exactly what that fuzzy, streaky haze is? Nice shot!
Thanks! It's crusty and dirty, but I kind of like it this way. I did clean it up fairly well (another version) but liked the image less as it lost character.
The diagonal streaks are stacking-artifacts from all the frames; that's the direction/path the starfield was passing in front of the lens.
The large halo, I believe, is lens issues coupled with a dark lens vignette. I could prob use a gradient and darken it right out.
... and across the bottom, horizontally, is sensor noise from the d5500.
Thanks for the response, but from everything I have read, it is possible to do much better than I am getting without having to do a few hours each night over the course of a month. I have seen great shots done just in one night with only 15 frames. At least they are way better than I am getting. I plan to go further away from the City light pollution and see if that helps.
So, just to understand correctly, you used sixty 1's exposures for that image? I may try doing 100+ images at 20 to 30 seconds with the tracker?
Yeah my attempt was 60 or 70, 1s shots aligned and stacked. I'd love a tracker. I'm also in a light pollution 9 zone. ugh..
My camera setup was pointing nearly straight up when Andromeda was directly overhead.
Yeah I'd try more shots at 30s each and see what happens.
there is a post in this forum that says andromeda from the garden (or something like that ) that is very inspiring and should help explain things
Thanks very much for the responses. I had read that thread, but wasn't sure it would help me much since it was done without a tracker. I am beginning to think that without a tracker you need to take tons (say 60 to 200) shots at 1 to 2 seconds, but with a tracker maybe 25 to 30 shots at 1 to 2 minutes. Is this a logical assumption? Also, since I am using a tracker, I was hoping that someone might be able to suggest why my image is so poor. There must be something I have done wrong, which is what I am hoping to find out.
Next clear night, I will try 50 to 100 shots at 30 seconds.
Can anyone say if the "light" shots I took are supposed to be so white and almost completely washed out? I will post an example of an unprocessed "light" file. This was taken with a D810, 300 f4 PF at ISO 2500 and 2 minutes. Is this what they should look like unstacked and unprocessed?
Yeah, that's overexposed. Are you shooting RAW or JPEG? Shoot RAW, and you'll get better results. Most if not all of the stacking programs handle DSLR RAW files.
I have only ever shot RAW with any DSLR. So yes, these are all RAW files. Can someone answer my question then on how all the people that say to shoot 2 to 3 minute exposures at ISO 1600 to 3200 and not get overexposed files?
It looks overexposed but many times it will look similar before processing. The only way to really know is to look at the histogram. If everything is pushed to the right and clipped off the right edge, you're overexposed and losing detail.
When in the field, you need to determine your exposure by taking a shot and looking at the histogram. I keep increasing exposure time or ISO until the histogram is pushed far to the right but not clipped. This is how you get the most light.
If you can't get 2 minutes, its likely because you have too much light pollution. You should be using a LiPo or UHC filter to cut it down, or better, get to darker skies.
But the key is to utilize your histogram functionality on your camera while in the field to find the longest exposure possible.
https://photographingspace.com/deep-sky-astrophotography-without-telescope/
check out this article as it looks like these guys maybe know what they are doing. Like you I am still learning and am hoping to get some satisfactory shots of the andromeda galaxy myself. good luck and post what you get next time. I will be shooting it myself in Sept and I will try to post what I get if it turns out.
i think that you are over exposed
Yeah, definitely overexposed, but then how are people able to do 3 minute exposures at f2.8 ISO 1600? That is what I see recommended a lot for M31. I don't think I have ever seen anyone recommending less than 1.5 minute exposures for high quality images of this subject. I guess I just need to try lots of different settings/options.
FYI, the galaxy just above and the fuzzy star-like patch next to the galaxy are both companion galaxies to Andromeda.
A couple of things, First, that's pretty good for a first attempt at M31. Much much better than my own 20 or so years ago.
To get the best calibration frames you should take them at roughly the same time as you take the light frames. This way the focus and temp are as close to the lights as possible. Focus and lens orientation is critical to remain the same as the lights for the flat frames, and you have to have the proper exposure for the flat frames, too light or too dark and they won't work properly.
the darks frames could be taken after the imaging session is complete. Just put the lens cap on and shoot about 10-20 frames of the same duration as your lights. Unless you have a cooled astro camera, you should try to take them at the same ISO and temperature as your light frames.
BIAS frames are the only ones you don't need to shoot at the same time. you can keep a library of them and reuse them from session to session. Put the Camera at the same ISO as your light frames and at the highest shutter speed and fire away (I typically have a library of a 100 or so Bias frames.
Try a program called Sequator for stacking. I sometimes get better results from that then DSS.
Thank you Craig. I did that with the calibration frames, but it did not help. Last night I went out and tried again, with high hopes, but I again FAILED miserably. So bad that I am about ready to give up on the idea. I did 100 frames at ISO 1250 and 15 seconds. I then tried 100 frames at ISO 1250 and 5 seconds. I then tried another 60 frames at ISO 1600 and 5 seconds. I stacked each session in DSS and got nothing. In every case they were MUCH worse than the previous time when I shot at ISO 2500 and 2 minute exposures, even though those ones were overexposed. I have to admit that I do not like to give up on things and have done very well at every other style of photography I have pursued. I tried stacking the light files with the dark,flat and bias files, and I also just tried stacking the light files alone. In no case was there any sign of the galaxy arms and all you could see was the bright galaxy center.
I am also now uncertain about my star tracker setup. Here's my process:
- level the tripod
- align the tracker to polaris in the location as instructed by my polar app.
- add the camera and lens and turn on the tracker.
- recheck the polar alignment to make sure it is still accurate.
- find andromeda and center it in the viewfinder.
- take a few test shots to see how they look and confirm its location in the frame.
- I use Exposure Delay Mode in the D810 at 2 seconds to avoid any chance of mirror slap or movement.
- take all the light frames.
After carefully setting it up 3 times last night, I am still getting slight star trails, even with the 5 second shots. If the tracker was working properly and set up right, the stars should be sharp with zero trails. So I am confused and frustrated with this process right from the first step. I have watched dozens of videos on how to set this up, but somehow somewhere I am failing with the whole process. I wish there was an Astrophotography club in town that I could join to get some help.
I ordered a LiPo filter yesterday to hopefully combat the light pollution, but I won't see that from the USA for at least 10 days.
Tim, don't give up hope. I've been doing astrophotography for 20 years now and it never gets any easier. I'm always finding new mistakes to make.
AP takes a lot of time to get right, in addition to the equipment failures, there's random crap that happens that you have no control over.
Can you post a couple of the raw images somewhere along with a couple of each of your calibration frames? I'd like to take a look at them.
If you're shooting from light polluted skies, you're going to have issues with exposure times. Stopping down the lens a couple of stops can help.
the darker skies you can shoot in teh better your images will be.
Thanks Craig. Sorry, but I dumped all the files since nothing was working. If I try this again my plan was to recreate all the files and try to do it properly from scratch. Although, I am having second thoughts now if I want to spend all this time and energy (read frustration) just to get one image. And M31 is supposed to be the easiest deep sky subject.
Hi Craig and thanks for the tip on Sequator. I have used it a few times now and it is giving me results as good as or better than DSS. Plus, it seems a bit easier to use. A question about it: Sequator just allows for the upload of what it calls "noise files". So I uploaded all 160 dark, flat and bias files. Is this correct, or does it just want the dark files? In other words, what is Sequator referring to as "noise files"? Regardless, I got pretty good results uploading all 160 calibration files.
Here's a shot of M31 from a couple of years ago that I processed in DSS. I don't have any details around t he capture such as exposure time and how many subs, but I do know it was captures with my Pentax K-1 and Explore Scientific 127 APO mounted on my Losmandy G-11, and guided via PHD.
The Pentax K-1 has the same sensor as the Nikon D810. Both are very low noise sensors.
Tim, ideally you should find a green zone or better, using www.lightpollutionmap.info
I would suggest the faster 70-200 at 200mm f/2.8 and you can crop the final image. This will also remove most coma in the corners from using the lens wide open.
DSS is to GIMP as Astro Pixel Processor or Pix Insight is to Photoshop... give the APP 30 day free trial a shot. Also don't forget your calibration frames. Minimum recommended is 50 bias, 15 flats and 15 darks. 100/30/30 is more ideal.
Your best data will be when M31 is at the Zenith, but above 50 degrees is excellent. Right now it is still early in the season for M31, but I would suggest capturing data from about 2:30AM until about 65 minutes before sunrise. Or at least try and get closer to 2 hours worth of data.
Your sub exposure length will depend how dark your skies are. Darker = longer subs. I set exposure so the histogram is between 1/4 or ideally 1/3 from the left. ISO 1,000-1,600 should be fine, but whatever you use, stick with it and bias/darks need to be the same. Flats can be ISO 100 using AV mode on the camera.
Tips from an Astrophotographer... bring painters (no residue) tape and once you get perfect focus, tape the lens. Also try Backyard Nikon for tethering to your D810. Last, get a Bahtinov Focus Mask and use Jupiter, Venus or Vega with ISO maxed and a couple sec exposure. Might also need a dew heater if humidity is above 85%
Image for attention, think it was 2.5 hours from dark skies at 2500' altitude.
Thank you very much Robert. That is helpful info. I think my first mistake is that I am starting at 10 PM and going to about 12:30 AM. So still fairly low in the sky and not really dark yet. This is going to be like shooting moths with many late nights! I have no idea what Backyard Nikon or a Bahtinov Focus Mask is, but I will look them up.
Can you tell me how long your individual exposures were for that image? It looks great and is the quality I am after.
A few more questions:
I see "temperature" being referenced a lot in creating these files. For me, temperature is basically white balance or image color temp. Are you guys referring to the actual temperature outside in degrees?
Craig says "you have to have the proper exposure for the flat frames, too light or too dark and they won't work properly" So how do you say what the proper exposure is for this? I used a white T shirt stretched over the front and pointed the lens at a morning sky, set at 0 EV. Shouldn't this be the "proper" exposure? If not, what is?
Can I assume you used an equatorial tracker for that image? Which one? Which lens?
I have been using Live View and zooming in on Arcturus for manual focus. Is this not good enough for focusing?
FWIW, I am using Photoshop (PS) after DSS and have used PS since the early 80's. I am quite knowledgeable with PS, so would there be any reason to get GIMP? Does GIMP do something for Astrophotography that PS doesn't? The only reason I am using DSS is to stack the files before processing in PS.
Tim, Here's a good primer on taking flats.
https://astrobackyard.com/how-to-take-flat-frames/#:~:text=There%20is%20....
Here's another guide to taking flat frames:
https://photographingspace.com/how-to-create-dslr-and-ccd-flat-frames-fo...
Temperature we refer to is the actual air temp so the sensor is the same temperature you took the light frames at. Not color temp/WB
a Bahtinov mask is a piece of plastic, wood, metal, etc that goes over the front of your lens/telescope and has slits cut into it. It creates a diffraction pattern that you can see move as you change focus. When the camera is in focus the moving diffraction spike is centered with in the other diffraction spikes.
Temperature, I usually take this as the sensors temperature or ambient of your lights so your darks can be at the same temperature, iso and duration.
Flat frames are easy, I use my MacBook Pro dimmed to 25-30% with a towel over the camera and laptop to block stray light. AV mode, ISO 100.
Equatorial mounts I use a NEQ6-Pro, CEM25p and SkyGuider Pro, depending on how much weight and complexity is involved. The SkyGuider is great up to 200mm, but can do 300mm carefully.
Live view is like tossing the dice, with experience you can get consistently close or almost perfect but the mask really lets you confirm perfect focus.
I was trying to say DSS = GIMP and APP & PI = PS. APP is what I recommend, much easier interface and really good to use if you are proficient in photoshop.
Feel free to call me if you want, DM for cell if you want more in depth answers
Thanks very much Robert
My plan right now is to give it one more night using all the info I have learned from here. I won't have my LiPo filter for awhile yet, so I plan to head out of my area to a darker "green" area. I'll see how I do next time and may take you up on your offer to talk on the phone. Where do you live?
I am fairly clear now on the dark and bias files, but still unsure on the flat files. Basically I did almost exactly what you do, except I did it in inside shooting at my 30" monitor with a plain white screen. Can I assume this will not work because it is not outside at the same temperature? So do you do it with your laptop outside at the same time you take the light files?
Lastly, what is APP & PI?
Flats the temperature, exposure and ISO does not need to match with your lights. BUT it is very important that your focus is the same. For this reason I always tape focus down and leave it on so the next day I can take my flats. It is also important that the illumination from the TV or laptop be as even as possible... you can use your TV just try to dim ambient light as much as possible. Ideally a semi dark room with TV or laptop screen. Just set iso to 100 and use AV mode, or manual is fine if you get the histogram to peak in the center or just a tad left of center.
Also, the moon is really bright right now. That is NOT good for DSO imaging. Wait a week if you can.
Flats correct cold pixels, dust motes and uneven illumination.
Bias corrects fixed pattern noise, and darks hot pixels / thermal noise.
Dithering makes a very significant impact in reducing noise as well. You can make a tiny movement with the RA buttons on the SkyGuider Pro every 10 min randomly back and forth to manually dither your light frames.
Dithering: https://www.youtube.com/watch?v=sPPctY-zQ9c
Some minor improvement. I was only able to capture about 30 minutes of exposure time tonight before the moon came up and washed things out. I still plan to head out to a green zone, but I was anxious to see if the help I had received here would move things in the right direction, While I still have a long way to go to get anything as awesome as your shot Robert, I do feel that with more time and a better (darker) location, I can get closer. Anyway, this shot is sixty-five light frames at ISO 1250 and 30 second exposures, with carefully collected bias, flat and dark files. I did 100/30/30 as you recommended.
The one thing I do notice is that this time it brought out many more stars in our galaxy. Almost too many for my tastes, as it detracts from the Andromeda Galaxy and makes it kind of busy.
It's a very nice shot! Do you still have the original fit file after stacking?? I would like to play with it and see if I can bring it out more. A lot of times the image will appear it's appear someone washed out like what you what you have but can be edited in post post. The more data you capture the better off your image will be. When zoom in you can see a little bit of star trailing.. The polar alignment was probably off just a hair. Polar alignment is critical for the longer shots. I might try 120 shots at 1 minute. 2 hours worth of data should give you a much better image. Please be patient as this is really a learning process of trial-and-error. Peter Zalinka has really good videos on YouTube As well as nebulaphotos. Remember some nights nothing will seem to go right, just keep trying.Other nights the universe falls into place and everything goes right.
Hey Tim, your limitations is how long you can expose without auto guiding. This is a factor of periodic error in the mount, weight of camera/lens, proper balancing, and precise polar alignment. Using a lens with a collar is really helpful for properly balancing on these small mounts.
The 300mm f/4 lens + a 30mm mini guide scope on the hot shoe would triple your exposure time, if polar alignment is good. You should be able to get 60s exposures at 200mm with the Tamron unguided, and 3 minutes guided. Id say probably 30-45 seconds unguided with the 300mm lens and 2+ minutes guided.
It is 100% worth traveling to darker skies... and ditch DSS. Astro Pixel Processor is far superior.
You can whatsapp me if you like at 1 904 248 9005
Cheers,
Rob