pipnina

Recent discussions about the repeatability and collimation of my scope have suggested that I likely suffer from my focuser not being tough enough to support the weight of my bulky mono imaging train. Seems rather cut and dry that something like a Baader ST (or other, I don't know of any alternatives to that esteemed unit) would be a big boon to my setup.

However I am also curious if mere payload capacity and low/no backlash are the only benefits. I noticed when moving from my 130P-DS to my PHOTON 8" F4, that the compression ring fittings held objects much firmer and truer than the simple bolts driven into the CC/Eyepiece. I am wondering if high end focusers improve on that again, and hold objects not just tightly and without damage, but also very repeatably.

For example, part of my issue with the collimation is that my laser itself is not perfectly collimated, and being the Baader Mk3 I cannot collimate it myself. So I must rotate the collimator or the thing holding it (that seems to be the best option) to work out the center of the circle. But if my focuser isn't QUITE holding it exactly the same every time, I have yet another error bar to contend with! Do these high end focusers help in that regard too?

Also curious how easy/hard people have found it to widen the hole for their focuser barrel when going from 50mm to 55, or even 75mm diameter sizes. Sheet metal like that isn't always easy to cut clean and could release pent up tension.

Also keen to learn if alternative high-end focusers (maybe in the 3" size) are well known like the Baader ST?

Cheers!

On 11/07/2022 at 22:39, ONIKKINEN said:

This one looks much better. I think this one had much better seeing also since your starsizes are much better compared to the focus test, but collimation looks better too for some reason. Top right corner appears to be the best according to Siril (pictured) and ASTAP image inspectors.

 

Star shapes are still a bit weird but that could also be in part due to guiding or something other than collimation (still guessing its mechanical, focuser most likely). But, i would say judging from the image that its still a bit out of collimation for some reason. But also i will admit that i would accept that as starsizes are pretty decent here and if binned x2 it would be hard to find something to complain about. Not sure how to proceed if i were you, but since your tube and mirror cell are good it leaves the sensor itself being tilted, focuser being unreliable or the mirror collimation itself as the causes. Or the collimation method you use is still somehow not quite ideal. Try the laser-through-CC method and see if something changes?

I gave the collimation a good re-do tonight as it was getting dark. I managed to pry the M48-M42 adapter from the CC (some of these threads get well and truly stuck!) so the 2-1.25 adapter that holds the laser would attach to the CC. It does make it a bit softer, but not a huge amount. I do think it made things a bit easier though so thanks!

As you can see I took a quick 5 second luminance image of a star-dense region and the FWHM measurements are MUCH better than before. But visually you can still see something isn't quite right when you look at the corners. Maybe I need to attempt barlow lasering to get it as bang on as possible?

Also I wish GSO made their mirror cell in the same way as skywatcher... My 130P-DS threadlocked studs into the mirror cell, then used thumbscrew nuts to pull the mirror or push it further away from the cell block, with three locking screws that drive into the cell. Never had any issues with this and it was SUPER easy and intuitive to make it work. This GSO one however seems like they put about 30 seconds of thought into it and went "good enough!". I have to both have enough tension on the locking screws to allow the adjustment screws to actually move the mirror, while also leaving it loose enough to move the adjustment screws without too much strain. It's just poor deisgn, and from an OTA that costs a lot more than the equivalently sized 200P-DS...

I think a focuser upgrade will be in my future too, I think it just makes sense to ensure it's all held properly since my kit is heavy and if it fits my corrector and collimator more snugly and consistently true as well, that can only make things easier. I found I had to double check my laser multiple passes over when rotating it 90, 180, 270 etc to check it was all in the right place... The focuser wasn't always keeping it quite the same way every time. When you changed your focuser, how easy was it and did you have to do any alignments with that, like shimming etc? Or do you just have to re-collimate?

Many thanks again!

9 hours ago, ONIKKINEN said:

Top right corner looks maybe OK in the LUM file. Bottom left stars are seagull shaped which means you are way out of collimation. With F/4 way out is not nearly as much as you'd think it is, mind you and it really needs to be almost perfect every time. Tilt inspector in Siril below.

 

I'd say figure out the mechanical side of things first before tackling possible sensor tilt and collimation. If you try to collimate a scope that is not mechanically sound you will just end up going in circles and it will never seem right because it really isn't possible.

@alacant usually comments that none of the cheap newtonians available today are out-of-box compatible with the demands of astrophotography in terms of mechanical stability and yours is probably not different. One more common issue you could have if the stiffness of your tube, which if not stiff enough can bend/buckle under load ever so slightly and so cause the scope to go out of collimation. Fitting a longer dovetail bar and spreading the rings out further apart from each other (i think it was about 50cm) while also tying up the top of the rings together with another dovetail bar will significantly improve stiffness of the tube. Not seeing to the mechanical side of things first will lead you on a wild goose chase where tools dont agree with each other and results vary from night to night.

Hmm, thinking about my setup when taking that image... I just put my telescope back to the position it was in.

It does seem like the weight of the camera/filter wheel/coma corrector would have been weighing roughly on the lower left part of the camera, but not quite the 45 degrees this image analysis would suggest.

I also took images last night of the bubble nebula, although those weren't focus tests and as such have much longer exposures, so I don't know if they'll be as easy to analyse accurately...

By eye, matters in this image don't seem as bad, but the objects are on opposite sides of the meridian, so one has the center of gravity of the camera pulling one way, and the other has it pulling in roughly the opposite direction in theory? What do you think?

NGC_7635_Light_002.fits

Edit: Alacant has stressed the importance of robust support, and no doubt at some point I will move things to a losmandy bar and puck, but the vixen style supplied with the scope is already rather long (skywatcher supplies 30cm I think, this one is much longer) and flipping the telescope from one side of the meridian to the other hasn't produced a shift in collimation on the laser since I followed one of Alacant's pieces of advice and removed the primary mirror clips and replaced them with neutral silicone sealant. So I am not sure tube flex can be blamed.

1 hour ago, ONIKKINEN said:

Yep, sounds very familiar all of that. The diffraction spike thingy is for sure a sign that there is some collimation or tilt issue, but its rather difficult to tell which one since both will have the same effect of taking the scope out of collimation. By the way, you mention tension and focus lock screws, do these move the focuser drawtube by any amount when you use them? If they do, that's a big problem. You can check just by inserting a laser and watching whether it moves across the primary and if it does, you gotta figure out a way to put them in one setting where everything works and moves and then never touch them again. I didn't figure out a way to do that with the stock VX8 focuser.

Try building something like this for the laser issue:

 

Its just a regular cheap collimation laser with the middle target section roughly 55mm from the coma corrector lens when this is threaded to the coma corrector. Easy to use, just thread whatever you have connected to the CC off (and everything behind it, including camera of course) and thread this in place of the camera+adapters. Mark the focuser position at infinity and you can now collimate the scope with the coma corrector in place. The laser dot will probably blow up in size a bit when it goes through the CC and back, but that might actually make it easier to see the centering (i found that to be the case).

Ah! If I tighten the locking screw it does shift very slightly, and as I undo the tension screw and it begins to go loose it does shift again, but doesn't appear to move visibly once it has some tension on it.

The laser also shifts position as I tighten it into the 1.25" adapter (this is one with only one screw, but it does have the brass ring that it compresses instead of just driving the screws into the thing being held.

here's some sample images from tonight, maybe the star images can give a clue?

O3 FOCUS TEST.fits HA FOCUS TEST.fits SII FOCUS TEST.fits LUM FOCUS TEST.fits

19 hours ago, ONIKKINEN said:

How is the focuser in your scope? If its not rock solid 100% reproducible with every accessory you have, you will simply not have good reproducibility in collimation. I thought my focuser was OK until i fitted a baader diamond steeltrack to my scope and found out just how much better than OK a good focuser can be (in truth the focuser i had was kinda bad). The laser and cheshire you used are, i would imagine, different lengths so they are at different focuser positions or just at different distances from the primary. If you have any side-to-side movement when racking the focuser in and out you will also have different collimation readouts at different focus positions so you will again be unable to get a result that agrees with some other method. You could try to eliminate the sideways movement but that might not be possible.

I find that i need to collimate with the laser at the focal plane and the focuser at the same position as a camera would be, and also most importantly, through the coma corrector. Collimation without the coma corrector is irrelevant if the corrector sits in the focuser when imaging, so i collimate through that with a laser. Try setting up a bunch of extenders and spacers with your laser to make it be basically a camera replacement and have the laser originate from the focal plane and see if you start getting some reproducibility.

I'll give changing the focus position a go for sure! I do think I have some sort of tilt or sag going on, if it isn't collimation being the issue. This is because I see my pairs of diffraction spikes looking parallel when slightly defocused going up-down the telescope tube (relative to sensor/focuser), but not parallel (converge on one side, diverge on the other) on the axis going up/down relative to the focuser (i.e. if the focus lock and tension thumbscrews are down, and the side opposite that is up.)

It's unfortunately very hard to get the laser to the exact focal point as the camera however since the focal plane of the telescope is so far away from the secondary, so much so that I have to withdraw the TS-GPU CC that I use by about 2cm from the focuser in order to achieve focus, with the focuser already outstretched as much as possible! But I can try moving the focus position up and down with the laser and cheshire in place to see what effect it has.

1 minute ago, markse68 said:

I used the wideangle lens that came with the ASI120- it captures the insides of the focuser tube to align to. I have an Ocal now which is basically the same thing but a bit easier to use as you have a nicer software interface and ability to remote adjust the focus- nice little tool.

Rotating the camera should tell you if it's sensor tilt or collimation I'd have thought yes

Mark

Ah, I have the 120 mini which doesn't come with that extra lens. I'll definitely give the laser rotation, and camera rotation a go however.

Now that it's getting dark, I'm going to see if my sealant has held the mirror as it should and re-attempt collimation

26 minutes ago, markse68 said:

I set my 8" f4 using my ASI120mm and circle overlays to get the secondary in the right place and now each time I use  it I check collimation with a laser pointer in a Paralizer 2"-1.25" adapter, rotating the laser to compensate any miscollimation of it or the holder to tweak the secondary adjusters to hit the primary donut, then stick a barlow on the laser to adjust the primary - I made a target a bit like the concentre with engraved rings that is jammed into the filter threads of the paralizer so I can see the projected donut ring and centre it. Has been working well for a while now though I'm a bit surprised it almost always needs minor tweaks.

Ae you sure it wasn't senosr tilt though causing the initial issue you noticed?

Mark

I have done my best to test for sensor tilt. I can't detect any but I don't have the full and proper jig to do it with. My best test was propping the baader coli laser on something so it was pointing at the camera's sensor, while the sensor was in a M42->1.25 adapter piece inside a 1.25" holder. I got the distinctive grid of dots and AR glass reflection, I tried my best to determine if the grid was rotating or not but it didn't seem to be.

I suppose the other way to test it is to flip the camera 90 degrees or 180 and take a second picture, and see if the pattern moves with it. However I managed to get round stars at all corners before, so can't that rule a tilted sensor out by itself?

The camera idea for secondary collimation is interesting. How do you get it to take a picture of the eye view down the focuser instead of the view a camera normally gets when at the focus position?

Sadly I don't have a barlow but I can try the rotation trick tonight too. Thanks!

I noticed two nights ago that my images had odd shaped stars in one corner, and a different shape in the opposite corner. I have managed to eliminate these issues before through good collimation however, so I know it's most likely not focuser sag.

Last night I tried to correct it by taking the camera out and putting my baader laser colli in. Problem with this is that if the laser isn't perfectly true to the focuser, the whole measurement is wrong. And the tolerances on all the bits are NOT good enough, even though I used that laser to get the good collimation I had during my iris nebula image only a few weeks ago.

So I collimate as best as I can with the laser again. I take an image, and the issue is unchanged. Ok, I put in my cheshire eyepiece, and it claims the collimation is WAAY off. So I collimate it to the cheshire eyepiece instead. I take another image, and it's actually WORSE!

So now I try with my cocenter. This is not really a primary alignment tool, but I found with my other telescope that if you shine a light down the aperture just right you can get a good feel for how the primary alignment looks. It thought it was perfect...

Then to make it worse, I tried a "reset" of the tensions and stresses of the adjustment knobs on the mirror cell... Only to find out that screwing in the adjustment knobs pushed the stud they drive, into the mirror, thereby pushing it off of the cork pads it was sealed to, making it unusable. I have re-sealed it today, and hope to retry it again tonight, but I just don't get how I am going to collimate this all properly.

I'm also really miffed that GSO, who made the mirror cell, decided on that stupid design when my much cheaper-per-inch skywatcher 130P-DS has a superior collimation control system, where the cell has the studs threadlocked in place, and the control knob pulls and pushes on the cell. I checked it only just today, and it makes a lot of sense that I'd find my old telescope so much easier to collimate...

I thought cheshire eyepieces were basically as foolproof as you could get, so I am now lost at how to get my F4 newt collimated well enough to achieve the round stars I have only rarely enjoyed with it thus far if anyone has any tips I'll gladly try them.

I am using the Ekos focuser module, which I have heard some people have issues with. I try to get the focus as close as possible manually and then run the routine, set to whole field HFR, but it can often detect noise instead of stars, or detect multiple stars instead of just the one that is actually there. Then I end up with donuts.

I've tried switching from gradient to SEP, I've tried adjusting the sigma, kernel size and had it run 3 samples per hfr measurement.

Do you guys have a preference for conditions to focus under? Like a single bright star, using short exposures? Or a fainter star field (i.e. in the plane of the MW) with longer exposures? Single star where you select one, or multi star or whole frame?

Maybe even another program to focus with (Ideally linux compatible), since I have heard people have issues with Ekos' focusing.

How do you also determine the backlash? I have a zwo EAF and a crayford focuser which has some good tension on it so in theory I shouldn't have much if any backlash? I am also not sure how to get a very good quality set of filter offsets, since I think from what other people have said, focusing on Luminance is the best start, then set offsets for all other filters relative to it?

Thanks

11 hours ago, wimvb said:

Quite correct. Guiding corrections can be up to 2 500 ms or 2.5 s, the default maximum pulse length. Besides, why would one use sub seconds exposures with a guide cam anyway? You'd be chasing the seeing.

Ah, this might make sense. I was wondering why even at 1s I was sometimes getting big delays between frames of almost 2s.

I mainly wondered if shorter exposures would help PHD correct for the oscilations better, I guess probably not.

6 hours ago, michael8554 said:

If you must use sub1 second exposures, use the PHD2 MultiStar mode, which compensates for "Chasing the Seeing".

THough I've not seen any posts where HEQ5-PRO users have to guide sub-second.

Michael

I have been using multi star mode til now, it does seem to work quite well.

20 hours ago, licho52 said:

Enable subsampling in the camera settings of PHD and see if that enables quicker fps.

Sadly with this turned on the calibration fails, so I think the feature doesn't work well when operated remotely. Thanks for letting me know it exists though!

5 minutes ago, licho52 said:

Enable subsampling in the camera settings of PHD and see if that enables quicker fps.

Is that the subframe tickbox in the bottom right? I'll give it a go! Thanks!

At first I thought I may have been running against limitations of the raspberry PI 4's processing or memory, but I recently switched to using my laptop (I7-6700HQ, 16GB ram, nvme SSD) and realised PHD wasn't *quite* performing as I had anticipated.

I'm using the ASI 120MM mini as a guide cam, which claims 30+ fps at its max resolution, although I anticipate that's set to JPEG compression and 8-bit mode. I have run my guiding on 8-bit raw mode by mistake before, which didn't actually improve the frame rate beyond the (slightly below) 1fps. I seem to be getting this 1fps update interval regardless of whether the exposure is set to 1s, 0.5 etc. Could it be an issue with my laptop, with a setting in PHD2, or maybe the camera really can't handle RAW frames at all above that framerate? The guide SNR went waaaay down in 8-bit raw mode so I can't imagine how poor it would be in jpeg-compressed mode!

My typical setup is to run PHD2 on the laptop, which is plugged in to the hardware directly, and then PHD2 reports back to Kstars indoors which is controlling the setup as a whole. In theory this should be ideal as guiding is not being performed over the network but the lag remains. I have tried to run guiding directly at the laptop with kstars running on the same machine but I still couldn't get 2fps guiding, and occasionally I get skipped frames!

Has anyone had this issue before?

I have a similar problem with my TS-PHOTON F4 200mm. Thankfully I am using the much longer TS-GPU coma corrector which I can pull out of the focuser barrel.
I have been considering asking my 3D-printer owning friend for a way to move my primary mirror cell about 4cm further away from the secondary, since this would solve the primary issue, as well as bring the camera closer to the secondary mirror, which means better illumination, and reduce the torque on the focuser.

I am worried 3D-printed plastic might not be tough enough for an 8" mirror+cell though.

12 hours ago, scotty38 said:

These should fit the normal focus side and when fitted the focus is locked solid and cannot be moved manually so there should be no need to be able to adjust focus tension anyway. That's on a refractor but I assume the principle is the same regardless.

This was the issue. I had the bracket on the wrong way around! I guess everything is easier after a good sleep and not being tired any more lol.

I got it all sorted now, and I don't put a huge amount of tension on the focuser, just enough to ensure the rotating metal shaft and the focuser barrel have good grip and the sag is corrected. It's still very fluid and doesn't take effort to adjust by hand, so hopefully the EAF should handle it all fine.

Gotta wait for another night to put it to the test though!

I waited 2 months for my ZWO EAF to arrive and now that it has, it doesn't fit, even with the 3D printed bracket FLO recommended to me!

I tried to fit the bracket any number of ways, but the only way I could think of that made sense was fitting the bracket as in the picture below, leaving the motor to connect to the fine control. However the bracket only barely fit on, and wouldn't fit at all if the focuser's cover was fitted. When finally in a position where it would be able to accept the motor, it was covering the threaded hole that lets me adjust the level of tightness between the roller and the focuser barrel! I need that applying some tension to stop the tube falling out let alone avoid sag.

So I changed strategy, and tried to install the motor using the supplied bracket and affix it to the focuser on the normal speed end. However this also occludes the tension and locking threaded holes, so I am none better off!

2 months of waiting and it seems I am still some way off of being auto-focused! : (

Anyone know what I should be doing? This is the TS 2" crayford focuser supplied with the TS-PHOTON 8" F4 Newtonian.

Image below shows how I tried to install the motor with default bracket, and also the 3D printed bracket in the orientation I tried to install it in before, for reference.

As usual, instructions would have been amazing from the bracket manufacturer but none are to be found!

Whether it's the cable setup, looking at black and white noisy pictures coming in on the laptop when focusing or taking flats, the live preview and graph in the guiding program, the grainy individual pictures we get, the temperamental nature of the equipment (always anticipating something going wrong with it!)... It almost seems like operating makeshift equipment in a sci-fi show. Me setting things up gives me mental images sometimes of being a real observatory starting observations because of the scientific-looking images I get when focusing. Sometimes it's more like The Doctor trying to set up something in a panic to defeat the Sontarans with cables running everywhere, lugging heavy equipment about, tapping around on the computer trying to get it all to behave etc.😄

Certainly whenever I try to speak to other people about it all they look at me as though I'm from Star Trek...

I have just completed a 3 day stint of imaging the iris nebula during these astro-dark devoid summer nights. The planets aligned and I got 3 nights in a row of clear skies, so I got red one the first, green on the second and blue on the last, due to my autofocuser having not arrived yet. It has come out quite nicely but I am missing a fair bit of data as setting my cam to cool at -10 or -20c caused the AR window to fog up after a certain amount of time, apparently in this 70%+ weather I can only do -5c at most. All in all I managed to get about 5 hours of exposure before the clouds and rain came back.

I'm quite pleased with the result, although I still have some work to do when it comes to either calibrating or background extracting properly, as I do have some odd colour gradients in the background that I had to fiddle around with post-RGB combination.

My guiding is still a little suspect in places, but having re-tightened the guidescope it did improve matters, so I suspect flexture is to blame. Either an OAG or a proper dovetail mounting for the guidescope+cam is in the future for this setup now.

I also re-processed to show the detail in the middle of the nebula better as well.

14 minutes ago, david_taurus83 said:

That looks like flexure. How are you guiding?

It is a guidescope so I suppose there is the potential for flexture.

I am using this scope https://www.firstlightoptics.com/finders/astro-essentials-50mm-guidescope-finderscope.html  In the default shoe of the main scope. It seems to work fine on some occasions and not so on others however. I've tried to get it affixed as tightly as is reasonable. On my last iris nebula attempt I was averaging a 0.35 second RMS, last night i was averaging about 0.8 on the same target.

30 minutes ago, Aramcheck said:

I don't use WBPP but the ImageCalibration process worked fine on the single file I tried. Any reason you haven't taken Darks or Bias frames?

Cheers
Ivor

 

I have read that bias frames aren't consistent in terms of read noise or linearity on CMOS sensors, and it's calibrated fine without them in the past. As for darks, I have tried to build a library of dark frames, but I always end up unable to isolate the camera fully from light (even in a room dark enough for me to need to sit down for a few minutes to see, camera under a tshirt, plugged with its rubber dust cap and socks pulled over the end, some light still got through!)

Maybe I can manually calibrate them this way, since it seems like the tool allows it to be done in batch! Thanks for the workaround I'll give it a go soon!

I've been using Pix for a short while now. And I've generally found it to do things a bit more hassle-free (if a little more complicated) than DSS, however today it has baffled me.

I collected my flats before the session started as usual, and then collected my lights as I went to bed. Today I try to stack the 9 good lights with my flats and I find Pixinsight will happily recognise that the lights and flats belong together, generates a master flat, but won't actually use it on my light frames!

As you can see in this comparison between single sub and stack, not only is the vignetting unchanged in shape but the dust mote remains unchanged as well.

The stacking process did not even generate a "calibrated" folder as I have seen it do in the past!

And in my WBPP screen it even recognises that the two belong together with the green tick! Yet even manually changing the flat setting in the top right from auto to manually select the flat frames, they still do not apply! I've even done this without any darks or bias or darkflats before and it's calibrated just fine.

I'm a bit lost at the moment, hopefully someone here knows what I'm doing wrong! : (

Light_003.fitsLight_002.fitsLight_001.fits

Flat_003.fitsFlat_002.fitsFlat_001.fits

I've attached the bare minimum Pix will want to perform a stack in case anyone would like a go

Many thanks!

NGC_7023_Light_003.fitsNGC_7023_Light_002.fitsNGC_7023_Light_001.fits

I seem to be having two other problems though now : (

My guiding seems to claim I have no peaks above 1.2 seconds, even checking back through the graph history. But my image keeps drifting in one direction and many images have doubled stars... Plus I suspect some dewing at the end of the webm videos.

Seems I still have problems to solve! At least the subs where guiding behaved seem to look pretty good now!

(edit, it claims the videos are corrupt, but you just need to right click and download them, I couldn't get Pixinsight's blink to work properly unless I chose webm)

I got the chance to test the telescope last night with the sealed mirror cell, and it seems to have worked just as you advertised! My laser collimator showed almost, or maybe even no movement swinging the telescope from one side to the other and my bahtinov mask diffraction patterns are now much cleaner and easier to read too! So perhaps poor collimation was upsetting it before.

Some comparisons of my defocused stars and bahtinov patterns:

Before:

After:

I was apprehensive to try modifying the scope as it would void the return policy but I think this has improved the scope so much already, if this is sustained I don't think I'll be re-evaluating this purchase any more.

When I get home to my PC which has the night's imaging on it I can show a proper subframe too, which I think it looking quite good. I seem to have a strange guiding issue however despite my guiding graph never showing spikes above 1.2 seconds-ish. Seeing doubled stars in some images, in different directions each frame.

6 hours ago, alacant said:

Here's hoping. The usual rule is: 

So you're sort of half the way there. But hey, at this game pragmatism rules.

EDIT:

Oh, almost forgot...

WARNING: the GSO cell.

You may have noticed that the black-knob mirror tilt screws can me adjusted so that they bear against the underside of the mirror (sic). You may want to put washers to prevent their clockwise motion allowing metal to glass contact. In fact I'd say it's essential. 

Do you reckon I should have used more? I could still apply some to the edge of the mirror where the side cork (I thought only the middle cork pieces needed to be sealed).

Now that it's sealed in, I suppose I would need to try and slip in some soft metal sheet or fabric from the under side to prevent the thread touching the glass? Is anything that's capable of preventing metal-glass contact sufficient or should some specific material work best?

I'll have a look into the springs although I seem to recall reading someone from FLO saying the newer GSO mirror cell springs are up-rated from the ones that famously needed replacing. Unless they're still wimpy even after that change?

Many thanks for the help so far.

I got some neutral "low modulus" sealant from screwfix yesterday and just got around to applying it to the mirror cell. Hopefully this is a sufficient application to hold the mirror, which is rather hefty. I did my best to avoid pushing it down while ensuring it was in straight. Tomorrow is pinned to be clear so I will hopefully be able to re-collimate and test it out now that I have not just done this, but also blackened the secondary mirror's edge.

After that, it'll be a matter of seeing if I can help the focuser sag at all...

Beautiful image! One of my favourite objects, and yet also one that I do not have the horizon to image myself (hill is in the way) : (

If you could remove the stars with starnet, then perform a wavelet sharpening (multiscale linear transform in Pix, or "contrast by detail level" in rawtherapee) on the star-removed image, it might bring out that little bit of extra sharpness in the fine details.