Creating a High Resolution Mosaic of the Milky Way Core

Over the last several years an ongoing goal of mine has been to create a high quality, high resolution mosaic of the core of our galaxy. Planning, executing, and completing a project of this nature requires a lot of time as well as a fairly high level of technical knowledge when it comes to equipment, acquisition techniques, and post processing techniques. In the last 5 years I've made various attempts at this kind of project, utilizing both full frame and APS-C sensor sized cameras and focal lengths ranging from 65mm up to 135mm. 

The above is one example, more of a mistake if you can believe it, shot in 2016 with a Nikon D600 and cheap Rokinon 85mm f1.4 on the original iOptron Skytracker mount. I honestly had little idea of what I was doing, I spontaneously decided to attempt a mosaic using the lens and camera while essentially testing various exposure lengths as a test of the mount at that focal length. As I began the mosaic I did a test of up to 5 minutes at 1 minute increments, doing 2 exposures at each, and ultimately continued that through 6 panels (2 rows of 3 columns). Afterwards in post processing I decided to try stacking each of the panels and really to my surprise, after essentially just blindly working through it all and stitching them, the result was quite fantastic. Then the frustrating portion happened, after my first initial editing attempt I was left with the above result, subsequent editing attempts could not replicate it and it seemed no matter what I did I was left with worse and worse quality. And so began the LONG 5 year journey to where I am today with this project. 

In the ensuing year, armed with a D800E and Rokinon 135mm as well as the new Sky Watcher Star Adventurer, I decided to really push the limits of my equipment and capabilities in this area and ended up with the following result:

At the time I was elated with the quality, the resolution was extraordinary and allowed for detail I had not really previously seen in a Milky Way core mosaic. While the result was fantastic the process introduced me to a number of technical limitations that really formed the foundation for the evolution of this project. First and foremost, the use of longer lenses and heavier camera bodies means not only needing a very precise polar alignment, but adaptation of more thought out weight balance mechanisms. At the time there weren't any fancy brackets, there were no tutorials on how to properly weight balance payloads on smaller mounts, everything about the ordeal was almost literally coming up with solutions on the spot via tinkering and testing. Creating a properly balanced declination load meant utilizing a macro rail Jerry-rigged to the mount, also including a ring to hold the lens (and reduce vibrations/flexure) which I made out of hose clamp because no scope rings existed that were of the size of the lens. Weight balance, I found out, was absolutely critical to getting anything longer than a minute or two with a keeper rate greater than ~75% on the mount at these focal lengths. 

This specific attempt also introduced me to the complexities of getting enough frame to frame overlap at longer focal lengths as well as the rotational issues that arise when doing a mosaic. On top of that the lens and camera combination, when using a light pollution filter, resulted in concentric ringing artifacts which could not be removed in post processing. Realistically several pieces of equipment critical to making this a realistic and efficient endeavor just did not exist at the time. There was little to no automation which made acquisition exhausting at best and often resulted in tossing out a lot of sub frames or completely missing a panel in some cases. Without access to a dual motorized mount with go-to, plate solving, easier weight balance and polar alignment features, not to mention camera powering and control, meant I was more or less spinning my wheels. This was more or less a crossroads in terms of continuing the project at all.

Fast forward a couple years, with the release of the Nikon Z7 as well as the availability of smaller, lighter lenses and a few more accessories for portable tracking mounts, I soldiered on and made another attempt at the mosaic utilizing a Voigtlander 65mm APO-lanthar lens. Both the camera and lens meant a large weight savings was possible, this made setting up the mount (still utilizing a Star Adventurer) much simpler for weight balance purposes, not to mention tracking capabilities and keeper percentages were a lot higher. The result was this:

Again, thrilled with the result (save for Jupiter being an ever present, constantly changing location object which made mosaic stitching problematic), and yet a number of recurring issues really took the wind out of my sails. The name of this game is efficiency and one problem continued to be the big limiter, a lack of automation meant an ever increasing demand on having to babysit the equipment AND resulted in things like reduced keeper rates, stitching problems, and more frustration. Other issues again reared their ugly head, locating and centering each panel, getting sufficient overlap between panels, having to sit there and make adjustments over time to achieve the mosaic. Another problem was the camera implementation of various firmware algorithms, this resulted again in concentric circles forming in the calibrated light frames and having to reshoot panels (not to mention various attempts at calibration frames). I quite simply needed some equipment that would make this a worthwhile endeavor, equipment that could solve a number of issues for me. Those equipment needs really boiled down to a few things solving the big picture headache:

Sky Watcher AZ-GTi:

The release of the AZ-GTi, a dual motorized axis portable tracking mount, which is essentially the bigger brother of the Star Adventurer. Initially this mount was intended for visual observation astronomy as an Alt-Az mount, it worked quite well for that, then someone suggested Sky Watcher implement firmware which would allow the mount to be operated in an equatorial mode. Game. Changer. Now we have a dual motorized axis mount that weighs all of 5 pounds, capable of being controlled via a smartphone app, capable of doing go-to operations. Also of note is that the mount power requirements are minimal, larger mounts require a dedicated power source such as a deep cycle battery for "portable" setups. 

Cooled Astro Cameras:

While various cooled astro cameras have existed for a number of years a number of new advances have made them a lot more affordable and reasonable to use in cases like this. Controlling and running the cameras is now simpler, newer sensor technology has made calibration easier, and powering the cameras can be done off of portable power sources (such as battery banks). These cameras also provide something basic that it seems is becoming harder to come by with consumer grade cameras: true RAW files, no firmware modifications of RAW data. 

ASIAir Pro:

The advent of the small Raspberry Pi based device has been a boon to mobile and portable astrophotography. Prior to these devices running complex astro equipment like dual axis mounts, cooled astro cameras, guiding, etc meant having to have some form of a computer with you. This meant having to power a computer out in the field on top of everything else. It also meant a lot of other things like constantly updating drivers, updating software, etc etc. Along comes the ASIAir, an all in one device that is small, lightweight, can be powered off a battery brick, and controlled by a smartphone or tablet. The beauty of it is in its simplicity, but that meant some limited features that are easily done by a standard computer. A key feature missing was automation of mosaics, being able to program multiple target locations to be ran in a single night with automatic slewing, plate solving/centering, and image acquisition. This was solved with the most recent firmware update. 

That equipment ultimately spawned an ultra-portable and easy to manage/power setup capable of doing highly complex mosaics of deep sky objects, so here we are today:

Sky Watcher AZ-GTi with ASI2600mc Pro, Voigtlander 110mm APO and ASIAir Pro