Eggy stars help

[Priesters]

Hi,

I think I know why I’ve got eggy stars on this image but thought I’d just check.  For this session my guiding RMS was averaging between 0.5 and 0.8” and I was taking 10 min subs.  I think my problem is that my imaging set up resolution was 0.47”/px and my guider resolution was 2.97”/px.  To my eye the eggs all seem to be pointing in the same direction which makes me think it’s a guiding issue rather than an optical one plus the fact I know I’m breaking the conventional wisdom of what constitutes a good set up.  However, would appreciate any thoughts on this.

Thanks

Andrew

[Borderline Bob]

They say beauty is in the eye of the beholder and I think this is a beautiful image.

[Priesters]

@Borderline Bob thank you, that’s much appreciated 👍.

[michael8554]

Hi Andrew

"my imaging set up resolution was 0.47”/px and my guider resolution was 2.97”/px. "

Seems reasonable to me.

If you want to know whether it's guiding or optics, you have to start with a single short sub, stretched if necessary, to eliminate guide errors.

Which of your scopes was this image taken with ?

Guessing from the pixel scale it's the SCT.

So it might be coma and a bit of tilt, looking at the image.

Michael

[Priesters]

4 minutes ago, michael8554 said:

Seems reasonable to me.

I’d read somewhere that there was a 4x rule i.e keep the guiding resolution within 4x the imaging resolution.  Interested to know if that’s not a thing and people successfully use set ups similar to mine.

 

7 minutes ago, michael8554 said:

Guessing from the pixel scale it's the SCT.

Yes, that’s right.  It could well be tilt or similar optical train problem as the attachment to focuser is not threaded.  
 

Next time out I’ll take a single short sub as you suggest and post it.  When you say short though, roughly how short do you mean?

thanks for the response btw 👍

[ollypenrice]

As I'm sure you know, you have no hope of resolving detail at 0.47"PP. The seeing will blur you out to two or three times lower resolution in long exposures. However, that blurring will not be directional. We'd expect it to give you larger stars but not eggy ones.

To my mind, the same would probably apply to a system guided at too coarse a pixel scale. Why would the error be greater in one axis than the other? Because that's where the greater error is?  Yes, maybe - but no more than maybe. A guide RMS of  0.5" to 0.8" should cover what the seeing will allow. The quick test is to double it, meaning your worst case scenario is to lose resolution after 1.6".  Are you resolving below 1.6"PP?

Have you plotted the long axis of the egginess you see against the orientation of RA and Dec? This is most important. If it doesn't align then, as already suggested, look to tilt.

And, after all that, where is this egginess? I struggle to see it. If you have to pixel peep to see it then your problem is... pixel peeping! Don't do it. Its only purpose in life is to spoil your enjoyment of a good image.

Olly

 

[Priesters]

52 minutes ago, ollypenrice said:

As I'm sure you know, you have no hope of resolving detail at 0.47"PP.

Yes I am aware of that.  I like shooting at F10 to get close up to small objects.  Technical question for you - if I used a reducer and cropped to get the same ‘close up’ image as the native focal length would there be any beneficial effect on clarity or would it be just the same?

 

1 hour ago, ollypenrice said:

And, after all that, where is this egginess? I struggle to see it. If you have to pixel peep to see it then your problem is... pixel peeping! Don't do it. Its only purpose in life is to spoil your enjoyment of a good image.

I shall definitely take heed of your very sage advice and stop pixel peeping.  

[ollypenrice]

11 hours ago, Priesters said:

Yes I am aware of that.  I like shooting at F10 to get close up to small objects.  Technical question for you - if I used a reducer and cropped to get the same ‘close up’ image as the native focal length would there be any beneficial effect on clarity or would it be just the same?

 

It is vital to get behind the phrase,  'Get up close to small objects,' because it has two meanings, one of them misleading and the other one valid.

Misleading: a long focal length allows a small object to fill the frame. It does, but with what does it fill it? If you are over-sampled (ie trying to resolve details which are lost to atmospheric turbulence even before they reach your telescope) then it fills the chip with mush. This is called 'empty resolution.'  Your pictured object is bigger but contains no more detail than it would from a shorter focal length (or bigger pixels.) It is no different from taking a small picture and resampling it upwards. In fact the latter would probably be better because the initial image would be likely to have more signal per pixel.

Valid: The closest you can get to an object is to image it at the maximum sampling rate that the seeing, guiding accuracy and lens-camera quality will allow. This has literally nothing to do with how big the object is in the frame. It is exclusively derived from sampling rate in arcseconds per pixel. What area of sky lands on each pixel? That's it. That's as close as you can get. So, yes, you'd be better off not oversampling in the first place and so get more light per pixel. With more signal the real detail you can extract in processing will go up. When it comes to presentation the best you can do is expose for long enough to get an S/N ratio good enough to present the image at full size (1 camera pixel given 1 screen pixel) and crop as necessary.

I used to shoot small objects in a 14 inch ODK with an FL of about 2.5 metres. I then switched to a TEC140 with a FL of a metre. I found no significant difference in resolution and the TEC was less troubled by very bright stars and never required any maintenance.

Olly

[Priesters]

Thanks Olly, very useful as always.  

At the risk of making myself look like a total idiot - is it possible to get to a more realistic sampling rate through binning and achieve a better outcome?