Our Beautiful Universe - Photos and Videos

[Drakkith]

I’m thinking of trying a 2nd round of focusing with the bahtinov mask after I get the telescope pointed at the target as that will give the telescope an extra hour or so to reach a stable temperature and also choose a target much closer to the zenith that way I’ll be shooting through less atmosphere.

What's the resolution of your system in arcsec/pixel?

 

[collinsmark]

NGC 5907, also called the Knife Edge Galaxy, or the Splinter Galaxy, captured from my back patio in May and early June, 2022.

NGC 5907 is spiral galaxy, seen edge-on. We don't see the spiral arms of this galaxy because it's like looking at the edge of a circular saw, we only see the sharp blade (of course the galaxy is not truly sharp, it's just made of stars, gas and dust, and is about 50 million light-years away). It can be seen from Earth in the constellation Draco.

Equipment:
Meade 10" LX200-ACF on an equatorial wedge.
Modified Starlight Instruments Feather Touch focuser.
Off-axis guiding.
ZWO RGB filter set.
Baader 3.5 nm Hα filter.
Optolong L-Pro filter.
ZWO ASI6200MM-Pro main camera.

Not seen in the image are tidal streams of stars swirling around NGC 5907, presumably due to a previous interaction with a dwarf galaxy. I have too much light pollution and not enough telescope to image the ghostly streams myself though.

Software:
Nighttime Imaging 'N' Astronomy (N.I.N.A.)
PHD2 Guiding (of course)
PixInsight
Gnu Image Manipulation Program (GIMP), only to combat a pesky dust mode on the L-Pro filter.
Topaz Labs Sharpen AI

A peculiar thing about NGC 5907 is, according to a study*, the galaxy has comparatively fewer giant stars, and those stars that are there are lacking in metalicity. This leads the researchers to conclude the galaxy primarily consists of lower mass, dwarf stars.
*(Liu, M. C.; Marleau, F. R.; Graham, J. R.; Charlot, S.; Sackett, P.; Zepf, S. E. [December 1998]. "Weighing the Stellar Content of NGC 5907's Dark Matter Halo". Bulletin of the American Astronomical Society. 30: 1258)

Integration:
Bortle class 7 (maybe 8) skies
All subframes binned 3×3
R: 42 × 5 min = 3.5 hrs
G: 52 × 5 min = 4.33 hrs
B: 40 × 5 min = 3.33 hrs
L-Pro: 89 × 5 min = 7.42 hrs
Hα: 57 × 10 min = 9.5 hrs
Total integration time: 28.08 hours

It's as if NGC 5907, like Snow White in the story of Snow White, after getting into a tussle with some unfortunate dwarf involving some white powder that got flung around, flies off in a rage, only to gather her gang of bloodthirsty dwarfs to get all stabby. Yeah, that galaxy's a real piece of work, I tell ya.

 

[Devin-M]

What's the resolution of your system in arcsec/pixel?

According to http://nova.astrometry.net/upload I'm getting 0.463 arcsec/pixel... I shot the Andromeda Galaxy a few months ago and and during that shoot I was getting stars down to about 16 pixels wide... this was about 2 hours of stacked 90 second exposures @ 6400iso with 2130mm f/14.2... no dark or flat calibration shots.

100% Crop:

 

[Devin-M]

Here's one of the 90 second @ 6400iso subframes:

100% crop:

Single Star:

 

[Devin-M]

Does anyone know whether the bahtinov mask needs to be in direct contact with the front corrector lens? Mine rests on the front edge of the tube so there's almost a 1 inch standoff between the bahtinov mask and the front corrector lens.

 

[Devin-M]

I’m trimming down a backup bahtinov mask I have laying around so I can put it in direct contact with the front lens…

 

[Andy Resnick]

I’m thinking of trying a 2nd round of focusing with the bahtinov mask after I get the telescope pointed at the target as that will give the telescope an extra hour or so to reach a stable temperature and also choose a target much closer to the zenith that way I’ll be shooting through less atmosphere.

Another possibility- I should of thought of this sooner- can you image Polaris with your setup? If so, then you remove tracking errors, you don't even need to have the mount powered on.

Never used a focussing mask, I don't know what is involved with that.

 

[Devin-M]

I’m hoping this will give me narrower diffraction spikes for finer focusing adjustments… almost an inch closer to the front lens…

Before:

After:

 

[Devin-M]

Another possibility- I should of thought of this sooner- can you image Polaris with your setup? If so, then you remove tracking errors, you don't even need to have the mount powered on.

Never used a focussing mask, I don't know what is involved with that.

I’ll try that tonight along with the new modified bahtinov mask.

 

[collinsmark]

Does anyone know whether the bahtinov mask needs to be in direct contact with the front corrector lens?

No. Keeping the Bahtinov mask somewhat forward (i.e., extended) from the front optics is just fine. It does not need to be in direct contact with the lens or the front optics or anything like that. As a matter of fact, I would be hesitant to put anything in direct contact with the optics due to risk of scratches and fingerprints. Just keep the center of the Bahtinov mask on the central axis. Other than that, it doesn't matter much.

I see I'm a little late in replying though.

 

[Drakkith]

My Bahtinov mask is 1-2 feet from the primary mirror of my 8" RC and it works just fine.

According to http://nova.astrometry.net/upload I'm getting 0.463 arcsec/pixel... I shot the Andromeda Galaxy a few months ago and and during that shoot I was getting stars down to about 16 pixels wide... this was about 2 hours of stacked 90 second exposures @ 6400iso with 2130mm f/14.2... no dark or flat calibration shots.

Hmm. That's about 7 arcsec per star, which seems a little high. However, you are imaging at f/14.2, so your airy disk minimum size is already upwards of 11 um, which might combine with other aberrations to make large star sizes.

If you're worried about focusing, check an exposure taken at the beginning, middle, and end of your imaging session and see how large the stars are in each one. If there's little to no difference, you're not having your focus shift in the middle of your imaging run.

 

[Devin-M]

Pointing at Polaris with a standard video panhead I was able to get dim stars down to about 16 pixels at 0.463 arcsec/pixel with a 15 second exposure at 3200iso (2175mm f/14.5 on a D800 36mp ff dslr)...

Polaris with its companion Polaris B at a distance of about 2400AU:

 

[Devin-M]

Comparison of bahtinov mask standoff distance from the front corrector lens (both 60 seconds at 6400iso)...

Direct Contact:

Approximately 1 inch standoff from the front corrector lens:

To my eye, the diffraction spikes on the direct contact option look a little narrower and "cleaner" at the base which could make it slightly easier to obtain fine focus...

Direct Contact:

~1 inch standoff:

 

[Devin-M]

I’m hoping this will give me narrower diffraction spikes for finer focusing adjustments… almost an inch closer to the front lens…

It's quite clear from this video the "direct contact" bahtinov mask has a narrower central diffraction spike:

 

[Andy Resnick]

Please see my answer to @collinsmark .

I am just curious that zooming in your first image yields something completely different than you 200% crop. Is it perhaps a matter of processing?

I understand.

Yes, your problem is 'undersampling',[...]

My last comment is that you are currently imaging at f/4.2 (if I did the calculation correctly). I'm somewhat surprised that your undersampling is so bad because when I image at f/2.8 (400mm) I don't get any undersampling- even when I image at f/1.4 (105mm) I don't get any undersampling effects.

Well, I spoke too soon... not the first time :)

This time of year, I remove my 2X tele and image Cygnus at 400mm (400/4), which is very comparable to timmdeeg's setup. Last night I got the following 'best' image (zoomed 400%):

Note- this is the 14-bit RAW that has been 'somehow' converted into a 16-bit/ch TIFF (and then to a jpg for posting). Undersampling effects are very obvious!

 

[Drakkith]

To my eye, the diffraction spikes on the direct contact option look a little narrower and "cleaner" at the base which could make it slightly easier to obtain fine focus...

I don't think you're having difficulty finding focus with an F/14 system. My 8" F/8 RC is MUCH easier than my F/5.4 80mm refractor, so your Mak should be pretty trivial. Especially with a bahtinov mask. The center spike doesn't need to be subpixel perfect. Just center it and move on.

 

[timmdeeg]

Well, I spoke too soon... not the first time :)

This time of year, I remove my 2X tele and image Cygnus at 400mm (400/4), which is very comparable to timmdeeg's setup. Last night I got the following 'best' image (zoomed 400%):

View attachment 303335

Note- this is the 14-bit RAW that has been 'somehow' converted into a 16-bit/ch TIFF (and then to a jpg for posting). Undersampling effects are very obvious!

So I'm no more alone. :)

Whereby I'm not fully clear what "undersampling" means. It's just the effect that the size of a star's image coincides with a four pixel square in my M13 image, what I could verify using a magnifying lens.
Now I hope for clear sky to do Dithering during taking the images and Drizzle while stacking as recommended here:

https://www.lightvortexastronomy.com/stacking-with-drizzle.html

 

[Drakkith]

Whereby I'm not fully clear what "undersampling" means. It's just the effect that the size of a star's image coincides with a four pixel square in my M13 image, what I could verify using a magnifying lens.

Non-technical, in the context of astrophotography, it means that your stars aren't 'round', they're blocky and highly pixelated.

Per high point scientific:

Undersampling occurs when the pixels on your camera sensor are too large for a given scope’s focal length. This creates blocky, pixelated stars. This indicates that there are not enough pixels within the star to create a round star shape. Oversampling, on the other hand, occurs when the camera’s pixels are too small for a given scope’s focal length. The incoming light is being spread over too many pixels resulting in a soft and bloated image.

For general astrophotography work, neither of these matter very much unless you're really picky about how your stars look. For planetary or lunar imaging using 'lucky imaging', you should always avoid undersampling in the sense that you should always choose a setup where your camera's pixels are smaller than the diffraction limited resolution of your scope, or about half the size of the airy disc. So if you're using an F/5 telescope, you should ideally choose a camera whose pixels are smaller than about 3.5 um, as the airy disc will be about 7 um. Larger pixels will cause you to lose detail that you otherwise could have gotten with smaller pixels.

When imaging deep sky objects, it would be ideal to match your pixel size with the maximum resolution provided by the combination of your telescope and the night sky, since the sky blurs the image in long duration exposures. But smaller pixels receive less light. So it's a trade off where larger pixels receive more light and require less exposure time, but smaller pixels potentially provide more detail at the expense of longer exposure times.

 

[timmdeeg]

Non-technical, in the context of astrophotography, it means that your stars aren't 'round', they're blocky and highly pixelated.

Per high point scientific:

Undersampling occurs when the pixels on your camera sensor are too large for a given scope’s focal length. This creates blocky, pixelated stars. This indicates that there are not enough pixels within the star to create a round star shape. Oversampling, on the other hand, occurs when the camera’s pixels are too small for a given scope’s focal length. The incoming light is being spread over too many pixels resulting in a soft and bloated image.

For general astrophotography work, neither of these matter very much unless you're really picky about how your stars look. For planetary or lunar imaging using 'lucky imaging', you should always avoid undersampling in the sense that you should always choose a setup where your camera's pixels are smaller than the diffraction limited resolution of your scope, or about half the size of the airy disc. So if you're using an F/5 telescope, you should ideally choose a camera whose pixels are smaller than about 3.5 um, as the airy disc will be about 7 um. Larger pixels will cause you to lose detail that you otherwise could have gotten with smaller pixels.

When imaging deep sky objects, it would be ideal to match your pixel size with the maximum resolution provided by the combination of your telescope and the night sky, since the sky blurs the image in long duration exposures. But smaller pixels receive less light. So it's a trade off where larger pixels receive more light and require less exposure time, but smaller pixels potentially provide more detail at the expense of longer exposure times.

Thanks for your comments, I fully agree and I don’t have a problem with wide-field images I want to capture. However I intend to see what happens after applying Dithering and Drizzle as recommended.

 

[Drakkith]

Thanks for your comments, I fully agree and I don’t have a problem with wide-field images I want to capture. However I intend to see what happens after applying Dithering and Drizzle as recommended.

I can't see I have much experience with either dithering or drizzle. I've used dithering a few times with some remote telescopes you can rent online, but only to move the image between exposures so that hot and cold pixels aren't always stacked on each other, which helps clean up the image when stacked since the extreme high and low pixel values can be easily eliminated.

 

[DennisN]

I'm reading this thread now sitting on the bus. I'm going to a new place a bit outside of the city. I hope it will be a bit better (Bortle 5 IIRC). The sky is crystal clear at the moment. I'm going to see if I can get at least some of the Milky Way with my fast wide-angle lens.

I hope I'll catch some stuff tonight, and if so, I'll post later in the thread.

Ok, here are some photos from that session earlier this April...

The site in general seemed pretty good for some stuff, but the exact location I chose was not good for trying to shoot the Milky Way (due to houses and trees).

Furthermore, the observation directions at the site were only promising to the west and south;
to the east and particularly to the north there was still a considerable light pollution "dome" from the nearby city (which can be seen in one of the photos below).

Even though I failed to shoot the Milky Way, I don't consider the visit a failure, since I now roughly know what that site can deliver.

Anyway, here are some wide-angle photos I shot at the site:

1. Starry night above the trees

Photo info:
Lens: Samyang 12mm f/2, ISO 1600, Exposure 5s, 163 lights (13.6 min total integration),
47 darks, stacked with Sequator and postprocessed in Photoshop.
Why are the trees so yellow? I'm not sure. It wasn't autumn, it was spring.
Maybe it's because they were lit up a bit by nearby streetlights?
And they stayed yellow because I wanted to express artistic freedom (i.e. I was lazy in postprocessing).

Detections by astrometry.net:

I noticed there was a bright cluster of stars close to the trees in the photo, and if I understand correctly it's the Beehive cluster (M44/NGC 2632) in the constellation Cancer:

2. Starry night over a lake

Photo info:
Lens: Samyang 12mm f/2, ISO 1600, Exposure 5s, 72 lights (6 min total integration), 47 darks, stacked with Sequator and postprocessed in Photoshop.

Shot looking straight north, where the light pollution "dome" from the nearby city is clearly visible. Considerable perspective distorsions are also seen here (at the horizon and in the lights in the water) due to the extreme wide-angle of the lens.

Detections by astrometry.net:

3. Starry night through the trees

Photo info:
Lens: Samyang 12mm f/2, ISO 1600, Exposure 5s, 76 lights (6.3 min total integration), 47 darks,
stacked with Sequator and postprocessed in Lightroom.

Look, the trees suddenly got green here! Did autumn suddenly turn to spring?
No, I edited away the yellow color heavily in Lightroom, which probably messed up the star colors too.
There are also some weird sky transitions in the photo; I'm not very experienced with Sequator yet (particularly not when mixing stacking stars together with a fixed foreground).

Detections by astrometry.net:

And last, this was the best shot I got of the Milky Way at that time:

Photo info:
Lens: I don't remember, ISO 800, Exposure: 1/8s (1/8s total integration).

(Camera used for all photos: Sony A6000)

 

[Andy Resnick]

Apologize in advance for the lengthy post...

Just finished imaging M57 for the year and am turning to various objects in and around the constellation Cygnus, starting with the Veil nebula. Here's M57, about 10 hours integration time (800/8, 8s subs), at 1:1-

there are a few features to note (which is why M57 is over to the side)- not just galaxies IC 1296, PGC2024204 and LEDA 2029852, but also the small asterisms near the center of this image- these are useful for gauging the angular resolution limit of my system (about 0.5 arcsec) and 'magnitude threshold' (magnitude 16 stars are visible).

Before moving on, I wanted to explain why I stop down my lens from f/5.6 to f/8: aberration control. It's most apparent in the corners- here's a 1:1 crop, from the upper corner of the (FX) frame, resulting from imaging at f/5.6:

I should point out that all things considered, the lens designers did a great job- this is a high-performing lens, and it's 40+ years old. Stopping down to f/8 gives this result:

Considerably better! The improvement diminishes as the image center is approached, but there is always better control of blooming.

Ok- moving on. I had a few nights where I could experiment using my 105/1.4 lens, again shooting at f/2 for aberration control. Tracking is less critical and so I can easily obtain 30s sub-exposures. Here's (almost) the full frame:

The color gradients are problematic, not sure where they are coming from. Honestly, I wished this lens performed a little better- upon close inspection, it appears that the lens image plane and sensor plane are slightly tilted with respect to each other, not sure if it's within spec or something requiring a repair. Here's M57 at 200% (no interpolation):

The color rendition is fantastic, but the lens performance falls off fast towards the edges, and not symmetrically (the lower left corner is by far the worst), coinciding with the large-scale color shift. On the other hand, in the central part of the frame (about a DX-sized region), the lens is fantastic- way outperforming the D810 sensor. Here's a single shot (30s exposure) at 800% so you can really see how well the lens does:

M57 is clearly resolved, color is great, and the stars have these weird PSFs that I think reflect the Bayer filter structure. According the APP, this frame has a star PSF FWHM of 2.5 pixels (2.48, if you must know), and I suspect the Bayer filter is large enough to interfere with the RAW-to-TIFF conversion.

However, I'm now turning to the Veil nebula (imaging at 400/4, 15s subs), I've only been able to acquire 30 minutes or so of exposure in 2 nights, but have been pretty happy with how initial results appear- here's a full-frame image with the stars removed:

This will only get better.

 

[DennisN]

Dew or dew not, there is no try

I was out training planet photography a couple of nights ago. It was definitely not optimal conditions (summer, city, planets quite low in the sky), but I didn't care since I wanted to train.

Lessons learned:

* Focusing on planets at 800mm focal length is darn difficult with a long lens mounted on a wobbly tripod head. And "darn difficult" is an understatement; it's more like maddening!

The tripod is excellent for normal photography, but it definitely reaches its limit in these kinds of situations. I am now seriously considering doing my pet build project - a motorized focusing mechanism controlled by a remote via a small cable.

The gear pointed at Jupiter in the early morning:

* Dew is now a problem for me too.
During the last part of the shoot I didn't notice it until I was starting to disassemble my gear:

Even though it was not optimal conditions I got pretty happy with the result anyway:

Jupiter and its four moons (composite)
from left to right: Callisto, Europa, Jupiter, Io and Ganymede

One fun thing is I could actually see in the original photos that the moons had different colors.
For instance Io was visibly more yellow than the others.

Photo info:

Jupiter shot at f/8, ISO 800, Exposure 1/100s, 20 photos prepared with PIPP and stacked with Autostakkert, and postprocessed in Photoshop.

Moons shot at f/8, ISO 1250, Exposure 1/5s, single photo, postprocessed in Photoshop.

Composition and edits made in Photoshop. Quite a lot of edits actually (chromatic aberration fix, noise reduction and some manual fixes).

Gear: Sony A6000, Tokina 400mm f/6.3 + 2x teleconverter (= 800 mm total focal length), Rollei intervalometer, K&F Concept tripod

 

[Drakkith]

Ah yes, the three brothers, nephews of Donald Duck: Dewey, Dewey, and Dewey. A very common problem during the nights. I keep a small blow dryer in my bag to help keep them away.

 

[Andy Resnick]

Another night of experimenting with the 105mm lens, this time using it as intended (large field of view) in the region around the "Coathangar constellation" (Brocchi's Cluster): 105/2, 30s subs, 3h total integration time

Post-processing this one (background flattening and subtraction) was difficult due to the combination of extremely dense star fields, large dust clouds, and vignetting caused by the large aperture- quantifying the background is a little tricky and I'm not sure how good a job I did.

 

[Oldman too]

ESO has, what must be, the coolest webcams on Earth. https://www.eso.org/public/outreach/webcams/
Check out the stars at night, you will be impressed with the cams.

 

[Oldman too]

Regarding those cams, I was looking around and came across something I don't recognize. Does anyone know what the purpose of these things are?

 

[collinsmark]

Regarding those cams, I was looking around and came across something I don't recognize. Does anyone know what the purpose of these things are?
View attachment 303690
View attachment 303691

'Took me a bit of googling, but apparently it's a 3-axis ultrasonic anemometers (wind sensors)
https://www.biral.com/product/ultrasonic-anemometer-3d-4-3830-20-340/#product-overview

Ultrasonic anemometers measure wind speed by detecting the difference in time taken for an ultrasonic pulse to travel in each direction between pairs of transducers caused by movement of the air. This method allows accurate measurement even at low wind speeds as there is no mechanical inertia to overcome. The lack of mechanical inertia also allows measurements to be taken at high frequency allowing precise time dependent measurement. Sonic temperature is also available in the output data.​

 

[Oldman too]

'Took me a bit of googling, but apparently it's a 3-axis ultrasonic anemometers (wind sensors)

Thanks, I was at a loss for search parameters. I noticed on some of the dishes that they had pairs, side by side but facing opposite directions.

 

[Oldman too]

Just curious, does anyone here use the MAST portal?
https://www.stsci.edu/contents/news...ence-data-after-the-end-of-commissioning.html

 

[collinsmark]

Here's my capture of Hoag's Object, from my back patio, 2022. We've discussed this object recently in posts #1693 through #1695 of this thread. Those posts inspired me to give it a try, myself.

Yeah, that ring in the middle is a galaxy. And it's about the same diameter of our own galaxy, roughly 100,000 light-years across. But in a nearly perfect ring. I mean, WTF.

Equipment:
Meade 10" LX200-ACF on an equatorial wedge.
Starlight Instruments FeatherTouch Crayford focuser modified for electronic focusing.
Off-axis guiding.
ZWO RGB filter set.
Optolong L-Pro filter.
Baader 3.5 nm Hα filter.
ZWO ASI6200MM-Pro main camera.

It's not my prettiest astrophoto, but it was kinda' difficult. The ring is too faint to be clearly visible in any given sub-frame. It took 20 hours of integration to pull it out of the noise. (The light pollution in my area had a lot to do with that.)

Software:
Nighttime Imaging 'N' Astronomy (N.I.N.A.)
PHD2 Guiding
PixInsight
Topaz Labs Denoise AI
Topaz Labs Sharpen AI

If I ever find myself with a lot less light pollution and/or a lot more telescope, I may consider revisiting this object.

Integration:
Bortle Class 7 (maybe 8) skies
All subframes binned 3×3
R: 16 × 5 min = 1.33 hrs
B: 54 × 5 min = 4.50 hrs
G: 24 × 5 min = 2.00 hrs
Hα: 7 × 10 min = 1.17 hrs
L-Pro: 134 × 5 min = 11.17 hrs
Total integration time: 20.17 hours

 

[DennisN]

Recently I was out shooting the Moon, and later when it was setting over the trees, I took a few shots with different exposure times to see if I could make something with them in Photoshop.

I could, but it took quite some work I hadn't done before*.

A cheesecake Moon setting

Photo info: Composite photo.
Camera and lens: Sony A6000 + Tokina 400mm f/6.3.
Moon details shot at f/8, ISO 1600, 1/200s exposure time.
Moonlight and foreground shot at f/8, ISO 1600, 1s exposure time.

*Edits: Lots . I worked with 3 layers in Photoshop (one for the background, one for the left part of the Moon and one for the right part of the Moon) and blending them together.
I also slightly increased the size of the Moon layers in the photo to get rid of all Moon overexposure in the background layer. Manual removal of some Moon fringes was also done. At last noise reduction in DXO Photolab and desaturation and increased contrast in Photoshop was done.And here's a stack of the Moon:
(I'm not sure if I quite nailed focus here, and furthermore the Moon was quite low on the horizon)

Photo info: 620 photos taken at f/8, ISO 1600 and 1/1000s exposure time.
Photos with quality 92% and higher (estimated by PIPP) were stacked.
Software: PIPP, Autostakkert and postprocessing in Photoshop.
Camera and lens: Sony A6000 + Tokina 400mm f/6.3 + 2x Teleconverter (=800mm total focal length).

 

[Drakkith]

Beautiful, Dennis!

 

[DennisN]

Beautiful, Dennis!

Thanks!

 

[Oldman too]

About that MAST portal question, I'm not familiar with the FITS or ECSV file format and could use some advice on working with them. I downloaded FITS Liberator 4 from https://noirlab.edu/public/products/applications/app001/ according to https://www.photographingspace.com/how-to-use-fits/ is the way to go. I'll be playing around with it for a while, mostly for format conversion and basic stuff. They also recommended using DeepSkyStacker along with the Liberator to take the edge off the learning curve for newbies (That would be me). I would appreciate any tips, advice or comments as I try and figure this stuff out.
Thanks, Scott