I need your feedback/help with my first DSO image.

[DMihai]

Hello everyone!

 

I would like your advice regarding a few issues I'm having with an image I took a few days ago. This is my first DSO image and while I'm happy with it, I'm sure there are a lot of things that can be improved.

 

The issues my inexperienced eyes can detect are:

•     The stars away from the center of the image are elongated, but they are not all pointing away or towards the center (which I understand would imply a wrong flattener – camera distance). This can be seen in the 2 light frames shared.
•     The vast majority of the stars are yellow. Could this be a processing issue? (I know there are a lot more yellow/white stars than blue, but still…). I was imaging at about 46 deg N and the target was between 60 and 80 deg Alt so I don't think the atmosphere was the reason.

 

You can find a light frame from each session, unprocessed stack and the final effort (jpg) here:
https://drive.google...f9v?usp=sharing

 

If needed, you can find the details regarding equipment/technique as an attachment so that I can keep the length of this post reasonable.

 

If there’s anything else that I forgot to add, please let me know.
Any feedback from you is greatly appreciated!

 

Thank you!

 

Cheers,
Mihai

Equipment-techique.txt

Edited May 25, 2021 by DMihai

[AstroMuni]

That looks beautiful. I dont use pixinsight but understand that there is a photometry tool in it that should help with star colours. Have you used that?

[vlaiv]

Hi and welcome to SGL.

Answer to the first question is probably - tilt.

When you have difference in corners like that it means that sensor is not perfectly perpendicular to optical axis. If star shapes change between first and last frame of the session - it means that you have flop that is causing tilt. It is best if you have threaded connection between components. Clamping connection can sometimes be less than rigid and as the scope tracks across the sky - gravity acts on camera and pulls it in different directions - that causes it to slightly tilt in different direction.

- solution to flop is solid threaded connection and removing flop/play from focuser if there is any

- solution to tilt (if removing flop does not help) is tilt plate

Answer to second questions is - yes stars are yellow - majority of them are. In fact, issue of color in astronomical images is not settled question and there has been a lot of debate about it. I personally think that there is a way to properly capture and render star color - other think that it is impossible (for some reason) to do so.

In fact - these should be star colors when viewed on computer screen:

There is much more orange / red stars then there are blue stars (yellow/orange make up 97% of all stars) - but are less bright, so when we image other galaxies - we can easily spot blue color in hot young stars as they are much brighter.

In any case - in order to properly render star color, one must take care to calibrate and handle color information properly.

[DMihai]

Thank you!

Yes, I color calibrated the stacked image, and the galaxy's colour seems close to "natural". That is, close to what others with much more experience are creating...

It seems to me that these are the colours the camera captured but there is something strange about it and I'm curious.

 

[vlaiv]

2 minutes ago, DMihai said:

That is, close to what others with much more experience are creating...

As far as color goes - don't really count on other images to provide you with clue of what accurate color is.

You can do a little experiment. Take celestial object - like a galaxy and do google image search. You'll find many different color renditions of the same target.

If you do similar experiment with known object - like can of beer that you are familiar with - you've seen it in person. 99% of images will show the same color. Those images were taken by random people - vast majority of which don't do photography seriously. Yet they manage to agree on color. Astrophotographers for some reason can't.

[DMihai]

Hello Vlaiv, glad to be here!

 

All the connections were threaded and I didn't notice any play in the focuser, though I did not specifically test for that.

 

Perhaps the adapter does not sit flush with the flattener or the camera and it introduces tilt. I am not convinced that it is tilt though. If you have the time, can you take a look at the 2 light frames please?

There is one light frame from each of the 2 sessions (2 nights).

 

As to the "true" colour, I agree with you. All I wanted to point out is that after the colour calibration the galaxy seemed "ok" but the stars less so.

 

 

[carastro]

That's extremely good for a first DSO image.  I noticed there are corners missing with an overlap of a different image below.  This suggests to me you have imaged this twice with different orientation and stacked the 2 together.  If there is elongation in irregular directions,   this could account for this especially if the tracking/guiding was not spot on, or there is tilt as Vlaiv suggests.

Other than that it is an excellent first DSO image, M101 is not an easy target I have found.  Meanwhile I will leave the mroe technical aspects to Vlaiv.

Carole  

[DMihai]

Hi Carole,

Thank you!

Indeed, the images were taken during 2 consecutive nights and have different orientation. However the elongation I'm talking about can be seen in the individual light frames and are less obvious in the final integration.

Looking into optical aberrations it seems that some of the stars' shapes are similar to those affected by astigmatism. Could that be the issue and not the spacing/tilt? Or could that be just another symptom of the spacing/tilt issue?

[vlaiv]

30 minutes ago, DMihai said:

Perhaps the adapter does not sit flush with the flattener or the camera and it introduces tilt. I am not convinced that it is tilt though. If you have the time, can you take a look at the 2 light frames please?

I already did that, but here, to be more precise, I'll give you a bit more detailed analysis.

First session frame shows just a bit of this effect.

Top left corner shows astigmatism:

With tangential component (one pointing toward the optical axis) dominating over sagittal component (perpendicular to first one - or along circle with center on optical axis).

Top right corner is similar:

with almost no sagittal component present

Bottom left corner is again similar:

Except, this time - sagittal component is dominant.

Bottom right corner is almost spot on (it also suffers some astigmatism - but I think that is probably feature of reducer / flattener at edge of this sensor).

3 corners astigmatic - two of the with tangential dominating - one with sagittal - and fourth corner almost perfect. That sort of implies tilt as tilt means that spacing will not be the same in each corner and "missed" spacing usually leads to astigmatism.

Second night image shows similar pattern, but this time - there is no winning corner - and corners change astigmatism orientation (here is image top left, top right, bottom right, bottom left):

sagittal, sagittal with strange center - a bit to the side, tangential, tangential with a bit strange center.

Now - it could be that there is no flop in the system that there is only tilt - and difference between corners is due to two different nights. If you disassembled your scope and put it back together next day and something changed in the mean time - like how you screwed things together - if some thread has double lead and you started threading with accessory 180° rotated or something.

If you left everything together and just put it on scope next day - then there might be flow. It is best if you examine first and last sub of the evening to see if there is flop in the system. If first and last frame have same/similar astigmatism in corners - then there is no flop - but if it changes between first and last frame (do take into account meridian flip - as it will rotate image by 180° - you need to look at same camera corners not same image corners) - then it is likely that something is loose.

By the way - here is handy tip for orienting camera the same way each session - orient it so that RA/DEC aligns with X and Y axis of the sensor (either "portrait" or "landscape" - whichever works better for the target). In order to do so - start exposure on bright star and slew the mount in RA at low speed (x1-x2 sidereal). Star should leave a trail in the image - orient the sensor until that trail is either horizontal or vertical.

 

[DMihai]

Thank you for taking the time to explain it to me in such detail! (And for the tips as well!)

Everything was disassembled at the end of the first session so the chances are that things were screwed back together differently.

I will do my best to correct any issues and hopefully it won't turn into an obsession and I will enjoy being under the stars.

 

Thanks again everyone, I wish you all the best!

Mihai