(Preface: Upon its completion, this treatise is much longer and far wordier than I had originally intended.  No doubt this is due in part to release of pent up frustrations after a long, wet and starless season, and part self-induced therapy.  Please indulge me, and perhaps you'll find some useful information or distracting amusement in the following ...)

Introductory Soliloquy

After a lengthy season of cold, rainy and cloudy nights, the "winter of my discontent" as it were, I have emerged from a seemingly endless, starless nightmare into brief but delightful moments finally adequte for imaging, at least from the vantage point of my light-polluted Pacific Northwest backyard.  Here I present my latest acquisition and processing of M63, the Sunflower Galaxy, in Canes Venatici (CVn).  For both the sake of my own record and perhaps the enlightenment, enjoyment or at least scrutiny of others, I will detail some of the steps I've taken in this processing, given the challenges of astroimaging under the shining dome of the man-made catastrophe of my local night's sky.

I have formulated long term plans to some day move to a new location that at the very least has darker skies and perhaps annually logs more clear nights than my present home.  For the time being, however, as I am still in my working years, putting daughters through school, striving to build up retirement savings, etc., I am likely stuck here under suburban skies in a part of the country famous for misty rain (seemingly) without end.  However, thankfully, there are actual stretches of clear nights from time to time, and I do what I can to take full advantage of them!

Target Details

M63 has been well-documented elsewhere, but I'd like to at least remark upon how beautiful it is to me.  Its type is SAbc, a spiral galaxy with no discernible central bar and loosely wound arms.  It's called a "flocculent galaxy," having a puffy, woolly appearance reminiscent of sheep or, surprisingly, a sunflower.  It's about 29.3 Mly or 8.99 Mpc distant, really quite close to us relatively speaking.  It is noted to be in the M51 (Whirlpool Galaxy) Group, which I had forgotten.  In my reading, I found that in the near infrared, there do appear to be two main spiral arms within all that woolly appearance, extending outwards on the order of 13,000 ly from the nucleus.  The gaseous structure of the galaxy extends outward as far as 130,000 ly, however that is not fully present in my image.  The final image (E) is cropped, centered around the main target.

Background Details

One of the things that I enjoy most about finished images is scanning the background for even more distant galaxies or other objects of interest.  I encourage you to examine the original fuller FOV image and expand it to full resolution.  I also provide (B) the full FOV Luminance Image, (C) its Inverted Image and (D) Annotated LRGB image which aid in navigation and identification of background galaxies.  Take a tour of the background and look at the myriad of distant galaxies, some of them at least appearing to be collected into groups.  Notable ones include PGC45992 (9-12 Mpc distance) at the upper right, and PGC46386 (16-36 Mpc distance) at the lower left.  (One may notice that these two corner galaxies are colorless, and this is due to some acquisition issues I will explain below.)  Looking up some of the identified smaller, even more distant galaxies, I found some on the order of 300+ Mpc distance.  That's upwards of 1 Bly!  (And to think I could capture these from my light-polluted backyard in between towering Douglas firs!)

Acquisition Details

I finally got a stretch of clear nights from March 15 - 19, 2024, during which I targeted M63 for 300s exposures, cycling through RGB only.  I had it in mind to just process an RGB image.  I set up my TSRC10T (10" RC) with TSRCFlat3 1.00x flattener and my QHY600M camera for a fairly wide FOV, even at 2000 mm FL.  I had no guiding issues earlier in the first night while the target was East of the pier (this is OAG by the way), but after the Meridian flip, guiding struggled due to a lack of decent guide stars.  The second night I rotated the camera about in search of decent guide stars for both sides of the Meridian, and things went better.  I ended up with the color acquisitions listed on this web page.  However, upon processing, I was disappointment in how "anemic" the image looked.  I determined that I needed to shoot some luminance channel and fall back on my usual LRGB imaging.  But I would need to wait for the weather to clear again ...

Finally, from April 16 - 20, 2024, I was able to acquire a decent number of 300s L frames, although their quality varied quite a bit from night to night.  Having learned the lesson from the earlier RGB session the month before, I made sure to test out the rotation position of the camera before initiating the imaging with target on the East side.  With the camera positioned with decent guide stars towards the East, I also rotated it 180° to ensure there would be good stars after the Meridian flip toward the West.  I used Stellarium to assist with this as it provides a nice visual of the position of the guide camera FOV upon rotation about the main camera FOV.  I had previously measured the angular distance from main camera to imaging camera by focusing the main camera on a bright star, then slewing to place the bright star in the middle of the guide camera, and then doing the math to figure out how to configure the offset in Stellarium.

As a result of all this rotating, my final image is oriented with respect to the L frames for the detail, however the RGB frames are somewhat rotated from this position, hence my corners appear monochrome.  Just think of the middle of the frame as the Land of Oz in full Technicolor, whereas the two corners represent Kansas from the start and end of the famous movie.  And pay no attention to the man behind the telescope ...

Collimation and Hardware Details

For the April imaging, I had some collimation issues the first night, but managed to fix things during the daytime prior to the second night.  I had already done a fair job of collimation of my focuser (using a Howie Glatter laser) and secondary mirror (using a recently purchased TSCOLLIT), but I had been remiss on the details of the primary mirror adjustments.  During the daytime hours before the second night, I managed to get fairly decent-looking and fairly stable images of the bright star Capella with my Hydrogen-alpha filter, and I was able to use this to carefully fine tune my primary mirror.  Late that evening I was able to double check with other stars using the luminance filter, and it turned out my daytime administrations had been adequate.

On this topic, the original tip-tlit connector that came with the scope was unfortunately and unnecessarily designed to substantially eat into the back focus budget.  I was able to purchase a shorter connector directly from GSO (see below), and this helped matters.  Even with this change, however, I was unable to effectively use my preferred Moonlite focuser as the train of flattener to OAG, CW and camera is quite long.  I resorted to retrofitting my thin TCF-Leo focuser from my days of self-flagellation with a demonic Schmidt-Cassegrain scope, and this recovered enough back focus to make the whole thing work for me.  However, since this focuser is triangular-shaped for optimum use with the SC and to clear its primary knobs, I am left always blocking at least one of my tip-tilt screws.  Once again, I am addressing this issue with the purchase of yet new hardware, this time a secondary mirror focuser.  If I get this to work, I'll be sure to post about it.

Note: I still don't see this tip-tilt part listed on anybody's website, but it is GSO Part FF354.  In contrast, the original part on my scope took up an additional 25mm (1 inch) or so of back focus.  This new one fit onto the back plate of the 10" RC scope perfectly with the same screws and springs.



For those interested, here is a picture of my current hardware train.  The TCF-Leo is not ideal as it blocks at least one of the tip-tilt screws, but serves its purpose to give me the room I want for the manual rotator and other equipment.  As I said before, I am addressing all this with yet another configuration of new equipment, hopefully to implement shortly.



Processing Details

Now with my March RGB frames and April L frames, I proceeded to calibrate in PixInsight, throwing out the most egregious rejects.  Calibration for me always includes dark subtraction and flat framing.  I always follow up then with CosmeticCorrection.

The next step is one that I have found to aid in the later combination and background subtraction of the combined frames.  For each frame, I performed AutomaticBackground subtraction, using Order = 1 for linear subtraction only.  I also maintained the background level using the Normalize setting (I've found some of these settings to be named counter-intuitively).  I should also point out that this kind of step is best done with an Image Container in which I've learned to check "Replace Image", although the output directory is different than the source images directory.  Effectively, this actually creates new images (again, counter-intuitively) for each frame in the new directory, but PI does not produce an image on the screen with these settings.  Not doing all this before, I was getting dozens or hundreds of images popping up on the screen!  As I said before, given my obscenely bright background, I find an initial linear subtraction, but maintaining average background, to be a good way to help mitigate some of my gradient issues, particularly when I am spanning a large swathe of sky and performing a Meridian flip.  The gradients under these circumstances can vary considerably during the night.

Following this, I examined the frames per channel in SubframeSelector in which I employ any number of approval expressions, usually based on FWHM, but also rejecting too low or too high SNR.  I won't go into details on this step because for me it is quite literally different for each image.

After this came registration of the frames with StarAlignment.  As I mentioned above, I used a L channel frame as reference with its given rotation, and that placed the RGB channels at a small rotation relative to this.  (But I can live with it.)

ImageIntegration is a step that I confess to which I have not been paying close enough attention.  This time around, however, I found that pixel rejection is actually a great way to further mitigate not just background noise, but also some of the reflections I can get from all the glass of the optics (flattener, filters, camera window, errant reflective surfaces I have not dealt with as of yet, etc.)  Having a bright background tends to amplify these issues and can prevent me from drawing out the kind of detail I wish to show in the final image.  Mathematically, there's only so much I can do with this given that the sky background, even corrected for gradients and smoothed out, has a criminally high signal strength, and any galaxy details right at that signal strength or slightly higher simply blends into the background.  There's nothing I can do about that other than to quit my job in the city, toss my daughters out into the world to pay for their own schooling, move out to the country with my gear, hopefully with the wife and dogs in tow.  Barring those steps, however, I will continue with the long integration times and aforementioned processing steps ...

After this came a long litany of linear and non-linear processing steps.  Like many, I've come to use BlurXTerminator, NoiseXTerminator, sometimes StarXTerminator to process the stars and galaxy separately.  In the linear steps, I do apply DynamicBackgroundSubtraction, the whole time looking forward to moving away from the city when I won't need to use and abuse this tool in such unholy fashion.  Finally, in the non-linear stretch, I combined the LRGB and took many steps using CurvesTranformation.  At this point, rather than declaring the image as finished, I abandoned it, orphaned like all works of art, be they great, mediocre, poor or abysmal.

A Passing Asteroid

During the first night of Luminance acquisition, I caught transit of Main Belt Asteroid Short (5627).  This prompted me to have some fun creating a video of its motion.  I'll post that soon and provide cross links to it here.  No scientific knowledge was gained in this exercise, only some measure of fun in processing the frames in PI and learning to use ffmpeg to create the video file.

Concluding Remarks

I hope you've enjoyed my image and loquacious account of its acquisition and processing.  If you have any particular questions about the steps, the gear, etc., please feel free to ask me, and I'll be happy to go into more details as requested.  Until then, I hope you are getting some clear nights and having some fun under the stars.  There are far worse things we could be doing at night ...