Perseus Double Cluster redux

[oopfan]

It's one of my favorite clusters due to the mix of bright red and blue stars. The last time I visited it was two years ago with my CCD. This year I'm using CMOS.

My CCD image can be found at the following link. The diffraction spikes were intentionally caused by a 3D-printed Newtonian mask that I added in front of my refractor's objective lens. Forgive me but the focus is off a bit:
viewtopic.php?t=3396

My Atik 314E CCD has a Read Noise of 5.3e- therefore it requires longer exposures than my CMOS Altair 290M's 1.5e-: exposure time of 60 seconds versus 6 seconds.

The Total Integration Time is almost the same: 2 hours. Both images show stars down to magnitude 19 to 20. However, notice how the faint stars in the CMOS image are perfect pinpoints and not ellipses. That is due to the short 6-second exposure. That's without guiding, only the mount's Periodic Error Correction (PEC). The CCD also only uses PEC but I had to reject a substantial number of frames due to poor tracking. Compare that with the CMOS camera's 6-second frames; I was able to keep every one of them.

I'm looking forward to using this CMOS setup for galaxies and nebulae. I only wish that the camera was cooled. It's not much of a problem in the winter months but it gets difficult in the summertime. The camera is inexpensive compared to cooled models. I think I paid a little north of $200 four years ago. The only thing I don't like is the 16:9 aspect ratio. I prefer a square format.

The specifics:
William Optics 71mm f/5.9 achromat refractor (don't buy it due to poor focus in blue)
Altair 290M
Optolong Red and Green Filters
Red Lights: 600x6s
Green Lights: 600x6s
Darks: 50x6s
Flats: 100 each filter
Bias: 100x1ms
Gain: 200
Black Level: 50
Mono12
FITS
Stacking and Processing: Astro Pixel Processor v1.083.2 for MacOS

EDIT: Unless you have used the 290M camera, "Gain 200" means nothing unless you look up the specs or refer to the sensor analysis. "Gain 200" is the threshold value where the Read Noise suddenly drops off a cliff, for example from 3e- to 1.5e-. My advice to everyone is to use that value when imaging. It may be called something different depending on your camera model, so refer to your sensor analysis. Also, resist the temptation to increase the gain further. The only time I use high gain is when plate solving and polar alignment. Also, don't ever choose "Black Level" of zero. (It may be called "Offset" depending on your camera.) There are plenty of write-ups on the SC forum on how to choose a good value, but if you are in doubt select a value that is mid-range on the slider control.

Brian

[Menno555]

Nice one Tim.
It indeed shows the benefits of of short(er) exposures.
Question: the stars in the core of the left cluster are showing "streaks", starting horizontally on top and going to around 30 degrees lower. No critique, just wondering what they are? It's only there, all other stars in the image do not have that.

Menno

[oopfan]

Hi Menno,

It was due to a major obstruction in the optical train: a dead mosquito. I didn't know it was there until I took flats at the end of the session.

Brian

[Menno555]

Ah okay!
Only had those (and lots of other tiny bits) in de cooling ribs of my camera. Plus the occasional web wever in front of my corrector plate.

Menno

[oopfan]

Menno,

Usually, I image with the CCD and guide with the CMOS (as well as perform PA.) So the cameras are affixed to their respective scopes. However, in those rare cases where I image with the CMOS, I need to withdraw the camera to do PA. I think that is when the mosquito jammed up the works. It makes sense since I do PA at dusk and that is when mosquitos are most active.

Brian

[timh]

Hi Brian,

I like this. The CMOS short exposure picture seems better to me. Imaging star clusters seems to be a skill in its own right - Difficult to get really good impact and brightness contrats with the bright stars really standing out and to get the fainter cluster members not only smaller but also proportionally less bright. At the moment I try to do this with star masking and 'binarization' in PixInsight which sometimes gives a good effect. Overall though star clusters just seem to be one of those things somehow where old fashioned visual astronomy is somehow hard to beat.
Tim

[oopfan]

Hi Tim,

6s red and green subs work very well on star clusters with my equipment. I too enjoy the linearity of this image. I was highly encouraged going into post-processing that the image required only a minimal stretch. I've found that if I ever have to push the stretch then stars become fat and they lose color. The best outcome is getting by without much stretch. At 6s the frames have some raining noise but a 2-hour integration and a black level adjustment eliminates that.

If I were to attempt a galaxy or nebula, I would capture two series: RG at 6s for the stars and another RG at 20s for the galaxy. (From experience 20s subs eliminates raining noise.) I'd process the runs separately before combining. For the 20s run, I would first remove stars and then aggressively stretch if needed. For the 6s run, I would only use minimal stretch as I did here with the double cluster.

Thanks,
Brian

[timh]

Hi Brian,

I am going to make a point of trying again this year to better image some of the nicer star clusters. Yes - as you say - low stretch is key to nice stars but also would like to get some good colour - so I plan to see how it works with longer exposures for chrominance combined with luminance from shorter exposures - then perhaps adding onto a starless image to get the dust in M45?

Tim

[oopfan]

Hi Tim,

I'm looking forward to your M45!

I need to get onto the APP forum to see if they have plans to add "star masking". I don't know if PI has it, but Star Tools definitely does. In that case I would only capture a 20s run (for galaxies with starry background,) followed by a star mask, and then aggressive stretching to bring out the galaxy while protecting the stars (in theory.)

Brian