[RCC] first try with M27 - Forums

Topic starter

Sep 19, 2017

We finally have our system up and can smoothly acquire data.

AP 6inch F9
ST-402 ME camera
SG-4 Autoguider (no dithering, but will have dithering soon)
AP 1200
FLI DF2

Using MaximDL and CCDStack

http://astrob.in/312883/0/

This is the first real image set we've gotten (~28L exposures at 5min each, half as many for RGB 2x2). Good contrast but the stars are blobby and hard for me to tell if this is a focus issue. It doesn't seem to be CCDStack since "Star Snap" is generally just working. Link to dataset:
https://drive.google.com/open?id=0B9u3R9NUcWHPcnNyTnBkbUhUa1E

I've noticed that contiguous autofocus runs give different positions that vary within a range of 100 steps, though the "V" shape looks fine. I want to address the issue by comparing the autofocus positions to what I would get with a mask. Does anyone have some advice to improve the quality of the image? I It feels like I should be able to produce something that looks a bit better in the same exposure time.

Helpful Engaging

Pat Rodgers

Sep 20, 2017 · Edited: Sep 20, 2017

Hi Maneesh
I had a play with your data set in PI. I personally wouldn't bother with binning your rgb. If you look at the histogram in your Astrobin image you can see the peak is over to the left, which is clipping data. I had this happening when integrating the Luminance into the RGB. In the end I couldn't stretch the Luminance as far to remedy this.
I used to use Maxim for processing and had varying results with the colour balance. I had a look at my version DL5 earlier and you can set the RGB channels individually mine are set at R 1.00 G 0.88 B 1.28 I remember getting these figures off the net in some tutorial.
Anyway here's my go with your data. The RGB Lum and LRGB images.

http://www.astrobin.com/313024/C/

Keep at it and clear skies to you.
Pat Rodgers.

Helpful Supportive

Maneesh

Topic starter

Sep 21, 2017 · Edited: Sep 21, 2017

Pat, thank you for the generous amount of work you put into things. Good to see what someone else can do with the same data.

What I understand from your post is that I lost detail in the background by setting the contrast too high but the 2x2 binned RGBs made it difficult to avoid doing completely. Is that right?

My current exposure times for each filter are something like:
L 300s 1x1
R 480s 2x2
G 435s 2x2
B 660s 2x2

The times I use are the relative ratios for flats for each filter. If I move RGB to 1x1, would you increase the exposure times, or simply try to get as many 1x1 frames as possible with, perhaps, the same exposure?

Do you see anything problematic about the focus?

Many thanks and clear skies to you,
Maneesh

Well Written Respectful Supportive

xordi

Sep 21, 2017

Hi Maneesh, I'm afraid you are right, the focus is really not perfect, this deviation is hard to fix it in the processing.

Peter

Maneesh

Topic starter

Sep 21, 2017

Thanks for confirming my suspicion. My FLI DF2 seems to work ok, as I said, I get what appear to be perfectly fine V curves with autofocus but the optimal focus varies in a range of +- 70 steps (0.2" total travel, 2000 steps) even on contiguous runs. There is no temperature control on the DF2.

I'll try to pay more close attention to the focus next run, but is the result perhaps not surprising given my equipment and that I am not temperature controlling? Or should I generally be seeing optimal focuses in repeat runs that are consistent to within something more like +- 10 steps?

Well Written Engaging

Pat Rodgers

Sep 22, 2017

Hi Maneesh.
Hopefully I haven't put you off binning. If your happy imaging that way carry on. From my understanding of binning it's an aid for collecting the coloured data quicker, so you can actually decrease your exposure times and get the RGB quicker leaving more time to gather your luminance. Where all the detail comes from. A very basic notion of binning, I know.
I'd try a set of subs all 1x1 though, on M27 again and see for yourself if there's any difference to the quality of both ways for your set up. Where I live the amount of imaging time is very limited to our poor weather atm and if it is clear the moon is normally near to full as well. So for me any RGB is done 1x1 so I can still get some sort of image at the end of the night.
I think my trouble with integrating your data was a Pixinsight thing. As it's set up for my cameras results, noise readings etc, and I'm loathed to changing them

Keep imaging and enjoying it that's the main thing Maneesh.
Regards and clear skies to you.
Pat.

Respectful Supportive

Maneesh

Topic starter

Nov 5, 2017

We took a dataset at 1x1 binning (and dithering is working now) for RGB and went through the excellent PI tutorials at light vortex.
I think 1x1 for RGB really helped. My L images still have really bloaty stars though, I'll be trying acquiring with FocusMax soon.

I sort of stumbled my way through PI without too much in the way of noise reduction or wavelet xform, up to basic DBE mostly HistorgramTransforms. I could reduce the stars a little, but some clear ringing in there. This *seems* better to me:

http://www.astrobin.com/320256/

Charles Harris

Nov 5, 2017 · Edited: Nov 5, 2017

Hello,
- you shouldnt ever have to bin any camera for a bright object like this one - imo
- you can zoom in on some stars after a sub and check the smaller ones to see if they are solid or like donuts. this will tell you if you are out of focus until you get a focus system
- you have blooming as well in your photo. probably from too many filters applied or maybe the wrong one in processing…or maybe due to out of focus stars prior to processing
- tighten up the focus and you will be much more satisfied i think

- good start - enjoy the experience ..
Charles

Helpful Supportive