Sunday, August 11, 2013

M 31 from the South Dakota Star Party

New dew controller in hand (see last entry) I was ready for the South Dakota Star Party. I had a short list of targets--The Iris Nebula, the Lagoon Nebula, the Heart Nebula. The one chosen would depend on the sky conditions. So much for plans.

Thursday night was partly cloudy with heavy dew and fog. I stayed home. Friday night started with high clouds, but went mostly clear as twilight ended. The view to the Lagoon was full of light from Sioux Falls, which was only about 15 miles away. I decided on the northern sky. I did a bore-sight polar align and tried to start with M 103. I switched over to M31 because I wanted to image something prettier. I got everything set up and running, but the first few frames were bad because the autoguider was losing the guide star in the remaining cirrus. There was a bright star at the edge of the autoguider's FOV, so I used that instead.

I set up the camera control to shoot five lights through each of the LRGB filters and then loop. That was at about 11:30. Amazingly, the system continued to run unattended for almost five hours! Focus held steady all that time, and tracking remained solid on all but one frame.

In the end I had to toss 25 frames because of clouds, one for tracking, and a couple because satellites passed right across M 31. Here's the result
Messiers 31, 32, and 110
Click for a larger version. Visit Astrobin for image details and a full-scale version.

I know the cores are overexposed, and maybe I need to do something about the blue fringing on some of the stars, and maybe I left it all a little dark. But I think I like it. Thank goodness I didn't cut the imaging short to chase after another object.

Dew controller and image noise

Well over a year ago I began noticing bands of noise on my images. Here's an example:

Stretched Light Frame showing noise band
And here's a patch of it at full scale (look just below the nebulosity). The appearance of the band will vary from frame to frame, but it's always a horizontal strip of hash. Longer exposures have more bands. Here's what it looks like at full scale:

Noise band at full scale
When that band covers your target, it overlays a great deal of noise that is difficult to remove. ImagesPlus noise removal reduces it, but you can be left with a light band.

I could not figure out what the source was. At first I thought it might be AC line noise (I was running using AC power). Moving to battery power didn't get rid of it. Bad cables to the CCD were suggested by others, but lots of cable swapping didn't fix it. Maybe it was the CCD camera itself? Online searches didn't suggest the camera had any problems like that. I did see some implications of interference from other devices being powered by shared batteries, so I isolated the CCD. The noise remained.

This was an intermittent problem, however. I didn't tumble to the cause until last week at Jeffers, while I was imaging M20/21. Almost every frame was noisy. For some reason I turned off the dew controller, and like magic the noise was gone. I imaged M17 the next night with no dew control (it wasn't needed) and every frame was free of the noise.

A more directed online search led me to a comment about the controller I use, the Model VI:

The Culprit: Kendrick Model VI
I immediately launched into more research and was happy to learn that Kendrick made an RFI-free controller that was in my budget range, the Standard Dual-Channel Controller (SDCC). $114 and shipped free from Astronomics.

The Solution: Kendrick Standard Dual-Channel Controller
A night of imaging with the SDCC keeping everything dry produced images that were all RFI noise-free.
If you're imaging and seeing a lot of horizontal noise bands, it may be your dew controller.

Wednesday, August 7, 2013

Two images from last weekend: M17, M20/21; Processing workflow.

I'm REALLY pleased with how this image of M17 came out:

Messier 17 (Omega or Swan Nebula), click for full size
The previous night's image was not quite as successful--M20 and 21. More about why in the next blog entry.
M20 (Triffid) and M21
Both images are full-frame. For all the gory details of acquisition, look at my gallery on AstroBin. By way of a quick summary, both are LRGB images made with an SBIG ST8300M CCD, TeleVue 102 apo and autoguided CGEM mount under mag 6+ skies. Dark and bias frames only, no flats.

Imaging is SO much faster doing LRGB in the dark than it is to use narrowband filters under light-polluted skies. 

I realize that the colors in the M17 image may look a little muted to you, but I kind of like them this way. What's particularly pleasing about the image from my standpoint is three-fold: The stars are nice and round right out to the edge of the field, meaning that my field flattener is doing its job perfectly; The merging of the LRGB channels seems to be spot-on. And finally, guiding with the Orion mini-guider works great at the 700mm focal length of the TV102 to correct mount errors. I'll eventually try out the mini on my C925.

My processing workflow continues to evolve. These two images were made by ping-ponging back and forth between Images Plus and Photoshop:
  • Calibrate, align and stack using IP
  • Stretch using PS
  • Noise cleanup and star size reduction in IP
  • LRGB combine in PS using Annie's Actions
  • Color balance in IP
  • Masked sharpening in PS
Yes, I know that's a mess, and some will say that all can be done in IP or PS or Maxim or whatever. Good for you if you can function with only one software package, but at this point I can't. I'm using IP 4.5, so probably some of the changes that have gone into later versions would make it a better processing tool.