Help with the best settings to get the most of a basic setup

[Jasonb]

Hi there,

I'm dipping my toe into AP, basically trying to see if I can capture what I see through the EP. I have a basic setup, and I'm not planning to spending any money to change it or improve it, and to be quite blunt, I'm not looking for any advice regarding equipment to buy etc.! I'm purely trying to maximise what I'm getting out of what I have, and a lot of that seems to come down to Camera settings, Deep Sky Stacker settings, and further Processing! I'm mainly interesting in Clusters and DSOs, like Galaxies etc. But I'm not expecting miracles here. I can't *see* beautiful, colourful, spiral-armed Galaxies through my EP, so I don't expect to capture them via this setup either.

So, to start, I have a Skywatcher 150P Newtonian on an Alt-Az mount. I have a loan of a Canon M50 Mirrorless camera, and I have a T-Ring. So the camera is directly attached to the Telescope focuser, and I'm able to get focus. 

I've done a fair bit of reading about camera settings, and I'm shooting in RAW, I've turned off all the Camera Noise Reduction settings, and have White Balance on the Camera set to Day. I've tried ISO at 3200 and 6400 and am not sure which is better, or if I should go higher still. Using a calculator I found online (https://www.lonelyspeck.com/advanced-astrophotography-shutter-time-calculator/) I am shooting at 0.3s shutter speeds, as this means I'm not getting any Star Trails with my non-tracking mount. The aperture is basically set to Zero as once the lens is off the camera my Telescope becomes the lens. 

What I'm currently doing is I use an EP to find what I want (say the Leo Triplet) and then I connect the camera to the Scope. I use the Camera screen to position the stars I can see on it, zoom in on the screen to focus (my scope has a dual-speed focus) and then I zoom out again. I note where a couple of stars are on the screen, and then I press the shutter. I have the shutter set up to wait for a few seconds after I press it, and then take 10 frames. At 0.3s shutter, this doesn't take long! I then use the slow motion controls to slightly adjust the positions of the stars on the camera screen back to their starting point, and I press the shutter again. Recently I've been getting 80 frames this way, so basically 24 seconds of exposure in total, but clearly that can very easily be increased!

When that's done, I put the telescope cover on, and take another 20 frames with the exact same settings for dark frames, and then I change the Shutter speed to the fastest speed possible (1/4000) and leave everything else the same and take 50 frames for Bias. I reuse these Bias frames, having taken note of the ISO first.

So, that's the actual taking pictures part! Any advice around that? It seems obvious, but should I take more, like hundreds more, light frames, just to up that total exposure time? Any advice about ISO, should I go higher?

Next is DSS. Following a few posts online, here's what I do. I change the RAW Brightness setting to 2.0000 and then I add all the Lights, and register them. I always check the "Star Detection Threshold" and try to get as high as possible (like in the low hundreds if that is possible, but at least getting close to 100), without it being obviously artificially too high. Once the Lights are registered, I have a look at the list of them, choose the one with the best score, and mark that as the reference file. I then change the RAW brightness back to 1.0000 and I add the Darks and Bias, and register them, telling it *not* to register already registered frames (thereby ignoring the already registered Lights). Then I stack them, going with Intersection Mode, and any other recommended settings (which tend to be Kappa Sigma Clipping with my short exposures.). It's set to use the best 95% of the frames for stacking, but the best I've managed is 72 out of 80, with around 50 being more usual.

So, that's DSS and stacking, Any advice around that? Any settings/steps I'm missing, or stuff I shouldn't be doing that I am?

Finally, probably my biggest issue, processing after stacking. I say it's my biggest issue because I don't know what I'm doing here, and therefore haven't really done anything. I have played with the basic processing in DSS (again, following a tutorial) which basically sets a Luminance Curve and then I move the R/G/B levels around. Yes, that *does* make everything brighter, so I can see more stars, and even smudges of a couple of the Leo Triplet Galaxies, but the sky is also brighter too. I have had Photoshop, GIMP and Rawtherapee recommended to me by friends on here, but haven't tried them yet. A friend on here put one of my stacked images through Photoshop but she basically said that the image is too dark, and doesn't have enough light (the Histogram is a thin spike on the left hand side, with no real room to the left of it to stretch it that way for example). So, that suggests that I need to fix something earlier in the process (taking pictures or DSS) before I get to the processing stage. Again I wonder, what are my best options here. Is the answer to getting more light/data into my picture just to take a lot more light frames? 

That's about it, I'd appreciate any advice that's out there. As I said at the start, I *know* this a very basic AP setup, and I'm realistic about my results from it. I'm not looking to change my scope/mount/camera etc. I'm just trying to get to a point where the final picture I get is *good* for the setup I have. In other words, to wring the most out of this setup.

Any questions, please ask away, and if it would help at all for me to upload any of my light frames, or one of the DSS autosaved stacked files, let me know. Thanks..
 

[happy-kat]

Hi

ISO your camera might peak at 1600 ISO anything above might be just too noisy.

Given your short exposures I would start with bright globular clusters, galaxies can be pretty dim.

Raw is raw the white balance is referenced when camera LCD viewing or to jpg when saved as far as I've read

Possibly lose the darks and bias with a Canon they can add noise. If you use SIRIL (free) for stacking you can play with a synthetic bias using the numeric 1028 (from memory I think this is the value) though you may find a different number works better for your camera specifically.

GIMP s free so would be an ideal place to start for post processing your DSS output.

 

Edited April 23, 2023 by happy-kat

[Elp]

Good that you're trying with what you have. Your current main limiting factors will be the focal length you're imaging at and the fact it's on an alt az mount. This vastly reduces the exposure time you can do which you've kept to a minimum as a result.

As long as your stars remain round open clusters shouldn't be a problem. Something bright like Orion needs around 3-5 seconds, Andromeda around 10s. Then you need to take hundreds of images. Best to plug in an intervalometer and let it keep snapping away automatically unless it has one built in. The issue with short subs is you're limiting the probability a sensor pixel receives a photon of light, it's a bit like lucky imaging you're doing. People who do that take thousands to tens of thousands of images to resolve target detail and increase signal, though you've already mentioned what you're looking to image, star signal should not be an issue as long as the stars are fairly bright.

Your iso seems high, higher iso will increase the noise in your images, too high and it will affect registration. I've generally never gone beyond 1600 on a DSLR. In DSS star threshold can be set at any number, it doesn't have to be high at all as long as there's a few stars it can recognise per image, the smaller amount the faster it will process a large image set.

I've found using GIMP you can linear stretch (levels) much harder than in PS.

[rickwayne]

With anything but the very brightest targets, you will definitely want a lot more total integration time. As you probably know, folks doing DSO with traditional techniques reckon on hours of it.

Best ISO depends on the camera. Usually above a certain threshold, all higher ISOs do is multiply the digital output by some factor, which you can do yourself in software designed for the task.

Concur with the recommendation for SiRil over Deep Sky Stacker, which is getting pretty long in the tooth these days (ha, as if I should talk!).

In an astrophoto the background is usually the most numerous value, right? So that's what your thin peak is. Ideally you want to get enough exposure that the peak is clear of the left edge that is, no black clipping. Left of that peak is nothing interesting, but your detail is mixed into the right part of it (unless you're doing clusters).

[Jasonb]

Thanks for your replies... Sounds like I don't need to be going higher with ISO, and in fact could be going lower. And it also sounds like that a lot more shots are needed! I'll have a look at GIMP and Siril when I get a chance!