StarTools

Last year about this time I picked up my AT130EDT and 0.8x reducer and imaged M42. But I was plagued by a couple of light leaks - I think there's a thread in the "Trouble" forum. I hadn't used the reducer since, but put it back on to have another go at M42. Aaaand - the "light leak" was back, except I think it's reflection inside the reducer rather than a light leak. (I made a CN thread about my troubleshooting https://www.cloudynights.com/topic/9111 ... p=13268407)

So here's what I ended up with for now. 965x15s (4 hours) at high-conversion gain this time around. Why the 15s subs? I wanted to preserve the trapezium without resorting to having to mask in layers with different exposures. I'm fairly sure I swamped the low read noise with sky background even at just 15s. Unfortunately, I lose a bunch of the ability to stretch because of the internal reflections. This is what Wipe had to work with/on: I've processed a few times, always ending up with slight variations on colors and saturation.

All things considered, I'd call that an admirable 4-hour Orion.
Perhaps a wee tint of green in the blue stars, and overall could maybe use a little more acutance.

On the right side, in deep images there actually is stuff over there. In fact there are two in EDSI right now - Dance and Yuexiao. But, yeah I guess the odd circle/rectangle pattern you have here isn't dusty structure huh?

What's the third shape along the top in your flat testing in the CN thread -- OAG prism?

In any event, if these come from artificial sources like a leak or glancing reflections, I think they are baked in. So flats won't fix that up, and routine Wipe will not see it as LP gradient. It's too bad the sides are reversed like that. But still, you might be able to hammer at it with Wipe a little more, to see if that allows you to then tease more stretch from the data. I would select synthetic flats in the inverse vignetting mode and then bump up the falloff slider, likely needing at least 50-75 percent. Then maybe try different edge behavior options to see if that can top it off. That one is hit or miss but sometimes it can help if the dataset is going wrong in just the right way.

Ah yes - green. It's always the green, isn't it? I have played with more drastic green throttling - and apparently I should, if you're seeing too much green cast to the blue stars. I was trying to leave a touch of green in the core, because (I think) there should be a bit of teal there. I'll return to the 'nuke all green' philosophy

As for the possibility that Wipe is just seeing real nebulosity there, I also took 20 minutes of M44 after M42 wandered too close to the powerlines:
With the M42 I did play with Inverse Vignetting, but I preferred what I was getting without that but with a great big dose of Gradient Falloff + Grow Opposite Axis, but it's all just different degrees of suboptimal.

I think i have thesame issue with the reducer. See the tread on CN.....

All halo's left aside it is quite a nice M42, imho

Wondering if you came across this one? https://www.astrobin.com/forum/c/astrop ... -at130edt/

Now that's a full frame camera, but still kind of similar. And his circle artifact does "fit" a little better across his sensor.

They float a few theories, so possibly from the connections to curvature on the underside of the reducer.

Since yours was a used rig, well I don't know if things have gone through iterations, either the scope, focuser, or reducer. The current reducer is a V2 though. I'll let you chase that down with part numbers and whatnot, if useful.

I wonder if you can do some "light cone" bench testing inside. I remember doing that when I was testing what turned out to be an aperture circumference flaw with my 100ED. Shined light through it in both directions looking for where the limits were.

I went back to see when I last used the 0.8x (things are fine with the 1.0x flattener). It was August on the Propeller Nebula. Here's ~2 hours of OSC for the star colors - I don't see the artifacts
So I'm a bit perplexed. Maybe I'll change the reducer-sensor spacing? Running a bit low on motivation to keep pursuing this right now, as we're into the season of using only the 1.0x

( Like Freddy, I don't seem to notice any issues with narrowband)

So it becomes even more strange. Now you see me, now you don't kind of thing..; It looks sometimes halo artifacts are introduced. The thing is i never experienced it with an 1.0 flattener either. So anyhow i must be someting with the 0.8 reducer on some occasions, the question remains , what occasion..