After months spent building my observatory, I’m finally turning back to the results of my “photon collection” and finishing a post I started some time ago.

Recap of Part I (from 8 months ago)

The first post (from 15.02.2025) set up a multi-part processing study of the ultra-faint planetary nebula Hartl–Dengel–Weinberger 3 (HDW 3), finding out that very dim targets require a slower, more disciplined workflow. It detailed the dataset of capturing 16 h 50 min in total (Hα 8 h 15 min, O III 5 h 35 min, plus 1 h each in R, G, B)—and began with the “easy” piece: producing accurate RGB star colours.

A step-by-step PixInsight pipeline was documented for the stars: Blink, WBPP, AutoDBE (ABE/DBE), LinearFit, BlurXTerminator, HistogramTransformation, NoiseXTerminator, SPCC, and Colour Saturation. While denoising, a tiny elongated field galaxy as faint as only ~30″ was noticed and somehow catched my attention. It was identified via SetiAstro’s What’s in my Image, SIMBAD, and LEDA/NED as 2MFGC 2869 / PGC 2263356, with a listed redshift z ≈ 0.03111. Using Hubble-law scaling, the post converted that redshift to a distance of ≈133 Mpc (~434 million light years).

With a nod to the ever-inspiring Carl Sagan—he might have put it like this: “Almost beyond imagination: that with instruments in our backyards we can call into focus objects so remote their light set out before we learned to dream; at such scales, ‘distance’ is a provincial word, and we yield instead to the patient measure of light itself.”

HDW 3’s filamentary nebula processing

After more than eight hours in H-alpha and over five in O-III, you approach the raw frames with certain expectations. In the case of HDW 3, the first stretch was—admittedly—slightly underwhelming: the filamentary signal sat only a whisper above the background. The photons were there, though, waiting for careful post processing.

Patience Over Processing: HDW 3 on Hold

I tried to process the data very carefully—soft masks, gentle stretches, no “invented” detail—but in the end I had to stop. The stacks didn’t meet my expectations. Most of the signal sat just above the background, and the O-III veils were only hinted at. Any stronger stretch brought noise, micro-artifacts, and faint halos. Beyond a very mild render it started to feel like „interpretation“, not information—and that’s not what I want for this target.

The nights of January 13–18, 2025 were obviously not the greatest: changing transparency and seeing (possibly some haze) hurt the signal-to-noise ratio (SNR) for such a faint object. Also HDW 3 simply needs more total exposure time — more hours in H-alpha and more depth in O-III—so the filaments are not just „suggested“ but clearly recorded. I’m leaving the current in-between result here, even though it’s too aggressively stretched …
Maybe as a kind of marker and a reminder that patience matters 🙂 Once I have the extra data, I’ll start over and I’ll return to the filament processing in Part 3.