This image of Sh2-308, the Dolphin Head Nebula is taken from a mishmash of test data I acquired while tuning and modifying my 6" imaging Newt.

The past month has been mostly cloudy, so in the absence of new data, I've been going back over some left-over files from earlier in the year. During the first few months of the year, I set myself a project of getting a cheap and cheerful 150mm GSO Newtonian into shape for imaging. I collected quite a bit of incidental data while testing the Newt and its modifications, and one target I used frequently was Sh2-308.

I ended up with 15 hours of data for this target, but it was of wildly varying quality. During the capture process, I'd had my Newt in and out of its tube rings, modified the primary mirror cell, moved the focuser and secondary up the tube by 40mm, flocked the tube, masked off the edge of the primary mirror... all modifications that improved the OTA but played havoc with getting a consistent set of images.

This data may have been a mess, but it told the story of how the modifications were shaping up. The metrics for the later sessions were significantly better than that from the earlier sessions. At the time, I'd been happy to see this improvement, and hadn't worried too much about the image itself. I'd simply made a quick, unweighted integration, enough to say: "been there, done that, captured the Dolphin Head, and improved my OTA in the process". I then filed it away and entirely ignored it.

However, the past weekend gave me a chance to revisit the data and see if I could make something more from it. I weighted the subframes to favour the later, cleaner data, while still hopefully gaining some SNR advantage from the earlier sessions. This gave me a surprisingly decent integration, but with a mess of misaligned diffraction spikes. So I performed a separate stack of just the final two sessions, after the modifications were mostly complete, and used StarXTerminator to combine those stars with the original stack. I then colour-corrected these stars with a paltry nine frames of OSC data I'd captured during my testing. After a GHS stretch of the nebula data, I combined this together, then did some final tweaks and cleanup in Affinity.

The SNR is certainly not as good as could be achieved had I captured 15 hours of "clean" data, but I think I've ended up with a credible result, given what I was working with. I feel it was definitely worth the effort to revisit this data, and I'm glad that I hadn't simply deleted the individual lights once the testing was complete.