Imaging star clusters

[timh]

Following on from Brian's post on the double cluster thought that I'd have a go at some of the Cassiopeia star clusters.

For 'real' visual astronomy (remember that?) I always found star clusters to be some of the most attractive objects-- and that imaging never quite seemed to do them justice. Still not quite sure what does work best or even if any general rules exist but having tried a range of exposures on different clusters am generally of the opinion that 1) the clusters that look the best are often the brightest having big coloured overexposed stars and Newtonian spikes and 2) a low stretch often works best so that the cluster itself is better differentiated from background and 3) there is not a lot of point imaging for longer than 20 min or so (at F4.5) - although a brief tour of Astrobin indicates that there is no consensus on any of this.

Anyway here are three ...the double double in the sword arm of Perseus, the dragonfly (or owl) cluster in Cassiopeia and Caroline 's (Herschell) rose,

Personally I think that the dragonfly is the most photogenic?

One curious question that I have. To my perception, stars in star clusters often look like beads on a string - often curved strings with dark circles and lanes between. It is quite reminiscent of - on a much larger scale -the way galaxies also seem to look like beads on curving strings (in that case presumptively along threads of dark matter). Does anyone know why stars might also appear to lie along threads in star clusters? Is that symmetry a natural consequence of the way radiation from one star igniting might help trigger ignition of a star close by within star-forming molecular clouds and thus end up forming a line of stars aligned with a ridge of higher molecular cloud density ?

------------------------------------------------------------------------------------------

SW PDS200 Newtonian down to F 4.5 (1.05 arcsec/ pixel)using the SW 0.9X reducer/ coma corrector, OSC AS1294 MC, 33s exposure at gain 124 frames all captured using Sharcap 4 and guided using PHD2 on a CEM70 mount. Bortle 6 skies with the moon up and 1 day past full.

NGC 869/ NGC 884 = 51 x 33s exposures

NGC 457 = 91 x 33s exposures

NGC7789 = 55 x 33s exposures

[oopfan]

Hi Tim,

"Beads on a string" - it's something I've noticed too.

I agree that NGC 457 (the "ET Cluster") is the most photogenic for two reasons: diffraction spikes and a pleasing mix of red and blue stars.

In the eyes of some astrophotographers, diffraction spikes are "forgivable" only if you have a Newtonian, however, for refractors it is a sin if you add them. Personally, I love diffraction spikes any way I can get them as long as they look natural.

In some ways, star clusters are more difficult to image than deep-sky. With deep-sky (i.e. galaxies and nebulae) you choose exposure based on light pollution, and then allow Integration Time to work its magic. However, with star clusters you choose exposure based on the dynamic range of the stars and the magnitude of the brightest star. Integration Time isn't much of a factor. You want the brightest star(s) to saturate to evoke diffraction spikes, however, a 12th magnitude star requires a longer exposure than a 6th magnitude star. At the same time you want to avoid over-saturation where you lose color information.

There is another reason why I enjoy imaging star clusters: The Moon is no longer the enemy.

Brian

[timh]

Hi Brian,

I think that the concensus (when I put the same question onto Astrobin) is that the beads on a string thing is mainly just perception. Although in reading more I found some interesting science on the non-random distribution of stars in open clusters ..https://www.astrobin.com/forum/c/astrop ... -a-string/

Yes I agree that clusters are actually quite difficult to image but the fact that they are 'moon - proof' is indeed an attraction. However one caveat to that. I thought that I would try longer exposures in an attempt to get more stars to show diffraction spikes. But of course that didn't work. All I did was end up with a much brighter background and histogram right over to the right. Then I realised that the only way that I was going to get more stars in the cluster to have diffraction spikes was to go for darker skies...hmm so back to no moon again?

Tim

[oopfan]

Hi Tim,

Hmm, "histogram right over to the right". Was the entire frame blown out or were you able to see stars? If you still saw stars, then LP could have been subtracted out, and hopefully you would end up with faint stars with diffraction spikes, theoretically.

Thanks.
Brian

[timh]

Hi Brian,

I also didn't image for a long enough total time probably (clouds came in quickly) but yes histogram not so far to the right as to be blown out completely. But what I observed was that- for the dragonfly cluster - using 33 or 110s exposures made no perceptible difference to which of the stars showed diffraction spikes and which didn't.

The diffraction spikes are obviously quite faint (and I am guessing that seeing or not seeing them adorning any individual star is an SNR thing - and that the darker the background the more stars would appear to show them? So rather than just increase exposure which helps a bit but seemingly not enough it would be more effective to decrease the brightness of the background sky -- i.e. surmising that detecting faint diffraction spikes (at say 0.1% or less of the total light of their cognate star) is just the same problem as trying to detect the fainter arms of galaxies?

Pretty certain that somewhere I must have some data to test this conjecture --- Bortle 4 and Bortle 6 Newtonian images of the same bit of starry sky- will post that if I can find it.

Tim

PS. Found it. Two images of the M109 region. Both just Sharpcap autostacks. Same set up. Both stacks of 45s exposures one at Bortle4 andthe other at my normal Bortle 6. I think that the former shows more and cleaner spikes and looks better?

[oopfan]

Hi Tim,

Yes, those Optics classes I took at university backs up your assertion. It's an SNR thing, not saturation.

Cheers,
Brian

[timh]

Just out of curiosity I did the dragonfly (or owl?) again tonight but this time without any moonlight. Exactly the same as above but this time only 55 x 33s. Just a sharpcap autostack processed in PixInsight using starmask and binarization to help suppress the smaller background stars. Think it is an improvement in terms of more impact..not sure..perhaps diminish the two bright foreground stars a tad?

Tim

[Menno555]

On the beads.
It's something called Pattern Recognition. Our brains are wired to see patterns, even when there are no real, physical patterns.
The propeller in M13 for example is a good example of that: it's just a random occurrence yet we "see" that shape in it.
The same for most "star beads": they are apparently in a line, so our brains make it a bead and beads are connected (at least, in our brain memory) and so we wonder if those stars are connected in some way.
A nice read about this here: https://en.wikipedia.org/wiki/Pattern_r ... sychology)

Menno