Absolute beginner - all pointers appreciated

[vineyard]

Hello,

I took my first ever subs of a DSO tonight.  20 subs of 20sec each of Maia in the Pleiades (taken with a UHC-S L booster filter - it's probably the totally wrong filter to have used tonight but I was just playing around with it!).  Stacked in SiriL.  Two files attached.  The first was the stacked image saved as a TIFF and then played around with in Lightroom.  The second was the stacked image manipulated a bit in SiriL first (adjusting the histogram) and then saved as TIFF and then played around with (less so) in Lightroom.  The second image has more "info" (ie fainter stars) which never showed up in the first one.  All tips welcome & gratefully received (whether processing or original capture or anything else!).

(I have the TIFF files or the original stacked FITS files too if anyone wants to break them open).

Cheers all,

Vin

[PeterCPC]

Difficult to make any meaningful comment without knowing the gear used but imaging is best avoided at times of almost full Moon (unless you are doing Ha).

Peter

[vineyard]

1 hour ago, PeterCPC said:

Difficult to make any meaningful comment without knowing the gear used but imaging is best avoided at times of almost full Moon (unless you are doing Ha).

Peter

Oops true, apologies Peter.  Yes you're right about the moon.  I just wanted to play with the kit initially to see if I could get the tracking working (the subs were a happy conclusion to the evening).

The kit is 102mm f/8.6 refractor, on a Vixen GP2 that I've modded with motors from EQStar EQ5 in Ukraine.  I can't do polar alignment b/c Polaris is behind a tall building, so I did a rough orientation w compass & latitude setting, and then calibrated using the moon.  Once the moon was centred, I parked the scope & reset that as home, and came back 2 hrs later, unparked it, and told it to go to the moon, and it found it again perfectly.

It was my first time using EKOS & the tracking so that was the main objective of the evening.  Told EKOS to go to the Pleiades - the astrometry module was amazing - it snapped the photo, did the matching and came back with the identification of Maia.  So then I thought I'd take some subs to practice the focusing & photo modules on EKOS.  And I was v pleased with how smoothly it seemed to go (don't have a motorised focuser so used the HFR graphs on EKOS).  No guiding, just tracking using the motors & EKOS.  Not bad tracking tbh - I could tell some slippage was happening (1 star that was on the first sub disappeared by the time of the 20th sub, I'm hoping a guidescope would improve that).  The scope was probably slightly unbalanced, and also since I'd only calibrated it with the moon 2 hrs ago I guess that's also inevitable?

I'm going to rummage with those Pleaides subs on Pixinsight given what Rusted managed to do with some moon subs I took earlier in the evening!

Cheers,

Vin

[vineyard]

Hello, so I restacked the original subs using master darks & biases that I had stacked themselves from 20 each (the original ones only used 1 random dark & bias).  And then I've played around on GIMP (never having used it before, Gina tipped me towards this as an open-source Mac friendly package to try) and so far this is where I get.

If I tilt the laptop screen, from some angles, the backgorund sky looks slightly greyish?  Is that high cloud (and the moonlight), or is it because I didn't take flats - and any reason why the grey is only on a smaller rectangle within the overall image?

Thank you!

Vin

[vlaiv]

Not sure what you are asking (either in original post, nor in this last one).

If you want to know what can be improved in your image / workflow - please post details and relevant .fits files. Maybe I've missed it, but I have no idea what camera you used to get above image for example (is it DSLR or dedicated astro camera. Is it cooled or non cooled model, etc ...).

If you are wondering about dark "frame" on right and bottom of this last image - my suspicion is that it is related to dark calibration, but I can't be sure until we examine master dark (and get details about cooling and such).

 

[vineyard]

Hi Vlaiv, oops sorry.  The camera was ASI178MC (uncooled).  20 lights of 20 secs each (w UHC-S filter in).  Similarly 20 each of darks & biases (but no flats).  It wasn't a proper imaging session (I was testing tracking) so I know there won't be huge data on the files, but I'm curious to see if I'm making some basic newbie errors & leaving info that could still be extracted just hanging there in the dark.  The fits files (for lights, darks & biases) are here.

Thinking a bit more I wonder if the "frame" on the right & bottom is from the tracking error.  I wasn't using guiding & hadn't done a super-accurate alignment and so the star that's an inch down from the top right corner was in the frame when the first light was taken but had slipped off by the 20th.  And so as the images get registered (if I'm using that term correctly) maybe there are only a few images of that border area, hence that seems darker as its undercaptured compared to the rest of the picture?

I don't know whether the haziness in the rest of the picture is clouds or b/c I didn't do flats though.

Cheers!

[vlaiv]

On 11/12/2019 at 16:32, vineyard said:

Hi Vlaiv, oops sorry.  The camera was ASI178MC (uncooled).  20 lights of 20 secs each (w UHC-S filter in).  Similarly 20 each of darks & biases (but no flats).  It wasn't a proper imaging session (I was testing tracking) so I know there won't be huge data on the files, but I'm curious to see if I'm making some basic newbie errors & leaving info that could still be extracted just hanging there in the dark.  The fits files (for lights, darks & biases) are here.

Thinking a bit more I wonder if the "frame" on the right & bottom is from the tracking error.  I wasn't using guiding & hadn't done a super-accurate alignment and so the star that's an inch down from the top right corner was in the frame when the first light was taken but had slipped off by the 20th.  And so as the images get registered (if I'm using that term correctly) maybe there are only a few images of that border area, hence that seems darker as its undercaptured compared to the rest of the picture?

I don't know whether the haziness in the rest of the picture is clouds or b/c I didn't do flats though.

Cheers!

Sorry I did not get back to you sooner - somehow I missed notification that you replied.

Here are my findings so far:

- above subs are not going to be much of use because they were shot in 8bit format. You need to shoot at 16bit format - so choose RAW16 as data format when taking subs.

- From fits headers I see that you used different software to capture your subs? Darks and bias have comment: "COMMENT Generated by INDI" while light subs have this: "PROGRAM = 'Siril v0.9.11'      / Software that created this HDU". Siril data does not include gain setting (darks and bias show that gain used was 145).

For calibration files it is important to use exact settings as light subs - so same gain, same offset.

There might be issue with offset, but I can't tell for sure because you used 8bit data format. There is strong histogram clipping to the left in your bias/dark subs - this can be due to offset issues but also due to 8bit format. What offset value did you use (driver settings)?

Bottom / right part of the image is definitively due to stacking artifacts - no subs contain anything that can be the cause of that, and there were enough drift over light subs to cause that much stacking artifact - you should crop it out.

Other tips would be - your tracking is rather poor. These are 20s subs, right?

Look at star shapes (I took one star and aligned it and made animated gif):

Almost every frame has some level of distortion. Maybe try to improve tracking / rigidity of your mount.

[wxsatuser]

On 10/12/2019 at 08:57, vineyard said:

The kit is 102mm f/8.6 refractor, on a Vixen GP2 that I've modded with motors from EQStar EQ5 in Ukraine.  I can't do polar alignment b/c Polaris is behind a tall building, so I did a rough orientation w compass & latitude setting, and then calibrated using the moon.  Once the moon was centred, I parked the scope & reset that as home, and came back 2 hrs later, unparked it, and told it to go to the moon, and it found it again perfectly.

Cheers,

Vin

Goto working does'nt mean the PA is good.

[vineyard]

Thanks both & especially for the detailed feedback vlaiv, much appreciated b/c it's exactly the type of advice I was hoping for.

I wasn't even aware it was shot in 8 bit, so now I know to be vigilant on that.

That's strange re the different software b/c I actually used the same software (the photo module of EKOS) and am fairly certain (as far as I can remember) left gain & offset unchanged.  I used SiriL for the processing & stacking.  I did register the lights in SiriL before stacking (b/c I knew the stars had shifted) but I didn't register the darks & biases before stacking them to make masters.  So I wonder if the reason SiriL is tagged on the lights (and doesn't show the gain) but not the others is because of that additional bit of file manipulation the software did?

I can't remember the offset setting.  Another lesson here: I'll probably keep a little journal of these settings so that I can see in the future.

Yes those were 20 sec lights, and agreed re the tracking being poor.  I only did a very rough alignment so I was expecting that (wanted to see how tolerant it is of user-error).  Also the scope was unbalanced and one of the motor belts can probably be replaced with something tighter (I think one belt slipped a cog a couple of times while slewing).  This was all just a first run of modified kit to see what the highest priority areas to start working on are - and this feedback has been v helpful for that - thank you!

Now just to wait for some clear skies to put into practice some of these changes 🧐

Cheers again,

Vin