Big stas round and tiny stars long ...

[John_Gill]

Hi,

I have been imaging Rho Orp with a Canon  DSLR piggy-backed on the scope and auto-guiding.  The settings on the camera were ISO 200, f4.5, @ 300seconds @ about 100mm.  After careful inspection of the lights, I notice that the tiny stars are slightly elongated but all the big stars are round.  Any explanation and or solution would be appreciated.

Look up
John

[Niall Saunders]

Hi John,

Well, first you would have to specify 'where' the apparent distorion was taking place.

In a 'perfect' optical train, every point-source of light entering thr primary apertuture (stars, for example, or ray-tracing sources in a hypothetical instrument, or even a model of a 'real' instrument) would come to perfect focus across the entire focal, or imaging plane. There would be no evident distortion.

But, in the real world, it becomes financially prohibitive (in most cases) to achieve this 'perfectly flat' image plane. Instead, optical systems present with 'abberations' that cause the plane to deviate from perfection - often by quite a considerable amount.

You might encounter radial distortion, where the image plane is only 'flat' at the centre of the image, and can deteriorate quite rapidly as 'radial distance' (from the image centre) increases - this results in 'coma', and stars seen through this type of system tend to become elongated 'tadpoles' in a radial direction.

You might also encounter lateral distortion, where the image plane is no longer perfectly orthogonal to the optical axis (as it should be for most OTAs). Instead the image plane is 'tilted' slightly in one of two mutually orthogonal axes, oftentimes it can be tilted in both of these axes at the same time, and by differing amounts on each. Now you get 'tadpoles' in each direction of tilt, or in the combined direction of tilt.

Add back in the fact that you may well also still have radial distortion, and you get what can be an almost undecipherable image, full of tadpoles seemingly pointing all over the place!!

This is where collimation of your optical train comes in. A well-designed OTA (designed for getting the best possible results for the investment made) will have the necessary adjustments made available to the end-user, and, inevitably - somewhere on the Internet - there will be a detailed discussion on how best to collimate that particular instrument.

Collimation will usually just not be possible on optical instruments not designed for this level of 'perfection'. As a result, nearly all of the distortions already mentioned (and perhaps a few more as well) will be present in every image acquired through the lens or mirror system.

I have spent more on a camera lens than I have spent on a high-end astronomy OTA - but that lens simply does not perform well enough to do much more than Lunar and Planetary imaging (and, even then, my ED80 at less than half the price, does a better job!!).

So, look at some of the tools available to you within PixInsight - especially in the Scripts section - and see which you could use to give you an idea of the distortion present in your image.
(Hint: a perfect image of a star-field would show 'equal' FWHM data for all the stars in the field, and every star would be 'circular' where the X- and Y-value of the FWHM would be the same).

[John_Gill]

Hey Niall,

Thanks for a very detailed explanation.  I was using a Celestron 8" EdgeHD scope and the camera was piggy-backed at about the 2 O'clock position.  The elongated tiny stars are across the entire image and run nnw to sse.  After extensive zooming into the image, I notice now that the big stars are also a little elongated (not as much as the tiny stars), so perhaps my auto-guiding and polar alignment should have been better .....

Look up
John

 

[Niall Saunders]

Hi Hohn,

If you want to exclude tracking and guiding issues - take some short (<30 second) exposures - and examine those.

Once you have that under control (or at least 'identified'), then look at Polar Alignment (whilst excluding guiding. Good, reliable, PA can make a huge difference to image quality - far more than might be expected.

Similarly, get your OTA perfectly balanced - in all three dimensions (if you don't know how to do this, then search for it and, if that doesn;t work, come and visit one of the Yahoo Groups that I post on where I discuss my approach to 'perfect 3D balance in 30 minutes').

With perfect PA and perfect balance, the guiding algorithm has a far easier job to do. And the end result - better stars!!

With your Edge HD, you have one primary point opf collimation (the secondary mirror adjustment) and most users assume that this is all that is needed. However, you need to be sure that your optical train (including focusers, focal reducers. camera mount rings. etc.) doesn't introduce errors after you have collimated the OTA. This is why the ultimate 'proof of concept' should always be a star-field test (on-focus, as well as off-focus - both 'in' and 'out'). If you don't have the means of adjusting the camera-to-OTA relationship, in terms of 'tilt', then you may need to consider that as a priority requirement.

I know this may all sound a bit off-topic for this Forum - but, after all, our ultimate goal is to improve our imaging - and, all over the worlf there appears to be an equivalent saying to, "You can't make a silk purse from a pig's ear!!"