Jump to content

Wizard Nebula - can't see where I'm going wrong.


Recommended Posts

Attached is a stack of 96x 300 sec light frames (darks, flats and dark flats included) on the Wizard nebula. It's 8 hours of data but I just can't seem to get anywhere with processing. Every time I get passed the linear stage, I just get a really mottled, ugly background. Are there any experts here who can tell me if there is something wrong with the data or if I'm doing something wrong?

Screenshot 2023-03-16 at 20.31.55.png

DSS Output - 2 Nights.tif

Link to comment
Share on other sites

47 minutes ago, CraigD1986 said:

who can tell me if there is something wrong with the data

Data does not support what you are trying to do.

Stars are too bloated for image to be used at full resolution. Besides that - it looks like there is some issue with optics or maybe atmospheric dispersion was too high - but channels don't seem to align properly. Stars are not round but rather "tear shaped".

image.png.d51b5c3665acc06a5aa4100dab009132.png

This is normal for ED doublet (I'm assuming image was taken with 72ED) - these scopes don't have the best correction and there will be some bloating in blue part of spectrum. However - stars should be concentric - meaning bloat should be around the core and not to one side.

Given that stars are as they are - when you remove them - you'll get very poor background as software tries to guess what is behind each of those bloated circles.

I'll try to process the data - the way I would process it, so you can see what can be accomplished (by me, I'm certain that others will have different approach and outcome).

Link to comment
Share on other sites

Thanks for your help. Is there anything I can do about this or avoid it in the future?

I tried to remove the stars when the data was still linear but this produced really strange looking images for both the star mask and the starless image that was left. (I tried Starnet 2 and Star Xterminator).

Also, If I try to stretch using GHS, the channels are way out of alignment. My intention was to use Bill Blanchan’s unlinked stretch and then HOO normalisation pixel math tools but these also produce funky results.

 

Link to comment
Share on other sites

result.jpg.25f2e7f94267accde21cbf7b0560e227.jpg

This is very busy star field and when stars are bloated like that - you can't really do much about it.

Also, don't attempt to stretch data further than it can handle - there will be too much noise.

18 minutes ago, CraigD1986 said:

Thanks for your help. Is there anything I can do about this or avoid it in the future?

Was the target too low in the sky? If not, we can rule out atmospheric dispersion, but then you should look into collimation of the scope.

Star bloat is easily resolved by using special luminance filter - like Astronomik L3 - that cuts a bit off ends of spectrum to prevent that bloating, but stars should be concentric for that.

If it is optics - it is either collimation of decentering of lens. I'm not entirely sure what is the cause, but tilt is not - it's something to do with lens itself.

Link to comment
Share on other sites

That’s concerning. I’ve never collimated the scope so it might be worth a look. I was using the L-enhance during this session. I can’t remember how high the target was but this data was shot in January of this year so I’ll take a look at the data and find out. 

Link to comment
Share on other sites

I also image using a 72ED, and my stars generally look like what vlaiv shows but only in the corners where I suffer with tilt, a problem I've spent far too long trying to fix and have never got a satisfactory outcome. So I'd be interested to see what is made of this.

What flattener are you using? I currently use a Stellamira 1x flattener with a DSLR, but I have just ordered the Skywatcher 0.85 reducer in the sales to see if it makes any difference.

Anyway, I don't have PI so I've just quickly run your image through Siril; background extraction, PCC, Asinh+300, Histogram stretch on default, Starnet removal, SCNR and GHS. The box shown is Siril's version of a tilt checker, and those values I can only dream of achieving with my 72ED.

Hope this helps, but I'm sure more experienced imagers will provide better help than I can 🙂

image.thumb.png.219bfd6a43527a0f7b1a12c56723cc40.png

Link to comment
Share on other sites

Naah; it's red! Seriously, on entrylevel equipment, data isn't always like as good as the hotshots on youtube. We'll have to live with it, and make the best out of it. On the way, we usually lear a lot. First, I'd say 300 seconds is a bit to long. The stars float out of shape, and in my experience, the bigger the number of subs, the less noise in the final image (as long as the signal isn't to low). This is processed using two free and open-source apps : The .tif is opened in Siril, and given a manual color corrction and a 'remove green filter', and finally a autostretch-bomb.  Then it's opened in Gimp, and processed in this order:

1 - Ian's Noise Reduction; the G'Mic plugin. Play with the options. Gamma all the way to the right. Best to do as early as possible.

2 - A wee furter stretch with levels (the bomb didn't detonate as it should). Take the left arrow all the way.

3 - A careful 'Value Propegate' on the stars, in order to supress them a bit. Select based on color, increase two pixels and feather three.

4 and finally - color adjustment and two rounds of a modest 'High Pass'.

My best tip: Reduce exposure time, and gather more subs. And lean a little heavier on Siril and Gimp.

DSS Output - 2 Nights_1.jpg

Link to comment
Share on other sites

The data isn't too bad, definitely an issue with CA or some sort of tilt causing misalignment of star colour, this is what I got with around 20 mins manual work:

WizardNebula-Copy.thumb.jpg.2d80af5b8c0a108bf565717207e4d67d.jpg

 

Starnet worked reasonably well too, I clone stamped out the harsh halos afterward and applied an NR:

Starnet.thumb.jpg.46324268fc8aa85642a4754aae257d9e.jpg

 

The key is to do only mild level stretching first on the linear data, then applying starnet or whatever software, harsher level editing can then be done on the starless version before combining back with the stars and final touches.

  • Like 2
Link to comment
Share on other sites

I have a sw 72ed and to get around my bloating of the stars  as Vlaiv says, I use an Astronomik L3 2inch filter with my scope and zwo asi294mc pro and really tightens the stars up a lot. 

Lee

Edited by AstroNebulee
Link to comment
Share on other sites

Thanks all for the input. I do own the Astronimik L3 but used the L-enhance as this was a hydrogen rich target. Maybe I’d also be better off using shorter exposures to control the stars a bit (maybe 120 or 180 seconds). If I got the same total exposure time but using shorter single subs, would I get the similar results to using 5 min subs?

  • Like 1
Link to comment
Share on other sites

17 minutes ago, CraigD1986 said:

If I got the same total exposure time but using shorter single subs, would I get the similar results to using 5 min subs?

Most likely better, only one way to find out. I shoot with DSLR, but have a few rules I stand by: I operate within a certain range of ISO (gain); 400 - 1600. The fainter target, the higher ISO. The more dominamt stars, the less ISO. When I've decided on a certain ISO I take test exposures. Finally I set on an exposure time where the histogram peak is within one-fourth and one-third from the left egde. Then I start gathering, and continue over several sessions untill I have at least 100. Experience so  far has tought me that much of the result lies in the number of frames, and the skill to drop frames before stacking. Cycle trough the subs before stacking, and watch the histogram peak wander back and fourth as you go up and down on the list. On closer inspection you will see that the most details lies in the darkest frames.

  • Like 1
Link to comment
Share on other sites

44 minutes ago, CraigD1986 said:

Thanks all for the input. I do own the Astronimik L3 but used the L-enhance as this was a hydrogen rich target. Maybe I’d also be better off using shorter exposures to control the stars a bit (maybe 120 or 180 seconds). If I got the same total exposure time but using shorter single subs, would I get the similar results to using 5 min subs?

With my set up the max exposure time I do is 180 secs. I'm currently processing my rosette image and the stars looks nice and tight and nice data for 2hrs 51 mins. 

Lee

  • Like 1
Link to comment
Share on other sites

1 hour ago, CraigD1986 said:

Thanks all for the input. I do own the Astronimik L3 but used the L-enhance as this was a hydrogen rich target. Maybe I’d also be better off using shorter exposures to control the stars a bit (maybe 120 or 180 seconds). If I got the same total exposure time but using shorter single subs, would I get the similar results to using 5 min subs?

 

39 minutes ago, AstroNebulee said:

With my set up the max exposure time I do is 180 secs. I'm currently processing my rosette image and the stars looks nice and tight and nice data for 2hrs 51 mins. 

Lee

That's interesting @AstroNebulee , I've just posted my Rosette at 2hrs 20mins, but using 300sec subs at ISO-400 with the lenhance (bortle 4). I'd be interested in your results with the same scope and a similar total time, but using shorter subs https://stargazerslounge.com/topic/407993-ngc2244-rosette-nebula-only-25-hours/

Generally I'm running 300second subs for my targets now, and always ISO-400...and still tilt issues!! 🤣 I'm less concerned with stars and more so about having fewer files (less storage and stacking time!), dithering and total time on target. Star size I can control with GIMP and Value Propagate, which I haven't done on this image admittedly. But I'm intrigued by the Astronomik L3 filter mentioned, it's not something I've come across previously so I might look into this.

  • Like 1
Link to comment
Share on other sites

13 hours ago, CraigD1986 said:

mottled, ugly background

Hi

The stack of the frames looks fine, there is nice detail in the wizard and the star field is spectacular, giving structure and interest to the shot.

Not a fan of PI, especially with the latest add-ons. It's easy to make it look over processed. 
Hardware wise, I think the focuser may need aligning with the optical axis of the lens. most 72s we've seen need this and is easy to do with a laser stuffed in the focuser.

My advice would be to keep it simple e.g. this is 5 minutes in StarTools. Even then i think I've overdone it!

Cheers and HTH

DSSOutput-2Nights-1.thumb.jpg.d1f22795cd81ccc40ac611e524102641.jpg

 

Edited by alacant
Link to comment
Share on other sites

As @alacant says there is some issues with the focus tube on the 72ed and needs alignment in some cases. I suffered with this too and here's the thread on it. 

I do use the rotator from RVO now and much better on helping with the tilt issue as less play in it. 

https://www.rothervalleyoptics.co.uk/rvo-m48-caa-360-rotator.html

 

1 hour ago, WolfieGlos said:

 

That's interesting @AstroNebulee , I've just posted my Rosette at 2hrs 20mins, but using 300sec subs at ISO-400 with the lenhance (bortle 4). I'd be interested in your results with the same scope and a similar total time, but using shorter subs https://stargazerslounge.com/topic/407993-ngc2244-rosette-nebula-only-25-hours/

Generally I'm running 300second subs for my targets now, and always ISO-400...and still tilt issues!! 🤣 I'm less concerned with stars and more so about having fewer files (less storage and stacking time!), dithering and total time on target. Star size I can control with GIMP and Value Propagate, which I haven't done on this image admittedly. But I'm intrigued by the Astronomik L3 filter mentioned, it's not something I've come across previously so I might look into this.

I will post my image as soon as I've processed it. I'm using a zwo asi294mc pro and gain 120 so may differ slightly to you as I guess your using a dslr. 

The L3 filter has been a revelation to my imaging and I love it and how my stars are controlled better. 

 

Lee

Edited by AstroNebulee
Link to comment
Share on other sites

Don't waste time - this is not sensor or focuser tilt.

Sensor or focuser tilt shows distinct pattern:

- there will be spot or rather "band" in the image where stars will be tight

- usually corners suffer and besides stars being slightly out of shape in some corners - they are also slightly out of focus.

In your data - you have teardrop shaped stars all over the frame equally.

Effect it the same as atmospheric dispersion - which separates wavelengths of light in one direction. If you want to understand how this effect might be formed in objective lens of a telescope - check out atmospheric dispersion corrector - which has two prisms and changes their relative position to cancel the effect of atmosphere.

You have two lens in your objective (that on small section act as prisms) and if those two are similarly displaced (or one is displaced with respect to the other) - you will get the same effect. You have 3 modes of displacement.

Tilt of one of lenses with respect to the other.

Change of spacing between the two.

Axis displacement (one lens being shifted "up"/"down" so that two axis are still parallel but don't align any more).

Out of these three - first and third can produce above effect. Second one usually produces increased spherical aberration (just bloated stars but still round).

Why don't you do a simple star test - take image of slightly in / out focus star - to see what sort of pattern it presents - and then we can know more.

Link to comment
Share on other sites

3 hours ago, CraigD1986 said:

Thanks all for the input. I do own the Astronimik L3 but used the L-enhance as this was a hydrogen rich target. Maybe I’d also be better off using shorter exposures to control the stars a bit (maybe 120 or 180 seconds). If I got the same total exposure time but using shorter single subs, would I get the similar results to using 5 min subs?

The L3 is purely a IR/UV cut filter, the L-enhance is for filtering your ionised gas lines in H and OIII so they are not doing the same thing. You should stack the filters. I have the L2 version and it is permanently fitted to my set up to avoid bloating.

Regarding the star shapes, I had this once and it was actually the tracking in RA that was shown to be the fault (I use ASI AIR btw), once I tweaked this I got my stars nice and round again.

  • Like 2
Link to comment
Share on other sites

14 minutes ago, bomberbaz said:

Regarding the star shapes, I had this once and it was actually the tracking in RA that was shown to be the fault (I use ASI AIR btw), once I tweaked this I got my stars nice and round again.

This is quite true - similar effect happens when tracking is affected, but it affects all wavelengths of light equally - so there is no separation of wavelengths like in above image where red is concentrated and tight and blue/green is bloated to one side.

Link to comment
Share on other sites

I had a chance to have a quick play around with this, data as mentioned seems pretty decent. 

So I used Siril for the main work and then tweaked in Gimp. I didn't go mad with the stars, and tried two versions with the lower version adopting a less is more approach.

wizarddsspostsiril.thumb.png.f314dbd0c8d7d166223a938ade23b8d5.png

wizarddsspostsirilless.thumb.png.8b45d68ec3da17a1eb2daa73da416f2a.png

 

  • Like 2
Link to comment
Share on other sites

Regarding your original question - have you tried to remove the stars after a stretch?

A 5 minutes play in PixInsight with the XTerminators + LHE and MLT can easily reveal some additional signal and details. Spending more time can improve the outcome, of course.

DSS Output - 2 Nights.jpg

  • Like 3
Link to comment
Share on other sites

39 minutes ago, alexbb said:

Regarding your original question - have you tried to remove the stars after a stretch?

A 5 minutes play in PixInsight with the XTerminators + LHE and MLT can easily reveal some additional signal and details. Spending more time can improve the outcome, of course.

DSS Output - 2 Nights.jpg

Have to say that is a pretty sweet image indeed. Well done.

It's the star processing side that really interests me in Pixinsight, although the entire package is rather compelling.

If I do get into imaging in a more meaningful  manner, I may indulge myself. 

  • Like 1
Link to comment
Share on other sites

Hi @CraigD1986I couldn't resist a process on your data

Here is my effort using Siril to remove green noise, background extraction, colour calibration, stretch it using auto then dialling it back a touch. Saving the image and using Starnett 2 on its own (not in Siril) to remove stars. Worked on the starless version in Photoshop by using curves and levels, bit of high pass and contrast boosts in certain areas. Used selective colour on the red channel and moved slider left to boost the reds. Slightly stretched the original linear image in photoshop to bring out the stars a bit, saved image. Loaded this in Startools then warped the stars to make them more round. Then combined the 2 (linear stars and starless) in photoshop and lastly Tpoaz Denoise the combined image. If I had a bit more time I think I could pull a bit more out of it. I did forget to increase the star colour. 

Sorry I rotated the image to just how I like seeing it and the wizard. 

Lee 

 

Craig SGL Wizard.jpg

Edited by AstroNebulee
Addition
  • Like 2
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.