Focus sesitivity

[Kaptain Klevtsov]

Just been working some stuff out to stew my brain some more.

My f/5 Newt has a light cone that starts at 200mm diameter and comes to a point after 1000mm at the CCD chip. The chip pixel size is 7 microns.

From the point of the light cone, for each mm I move up the cone, the diameter of the light cone is 1/5th of a mm, so that if I move 5mm, the circle is 1mm diameter.

If I want to focus the light cone on a single pixel, I need to get the focus within 35 microns, or about 1/30th of a mm. Is that correct, as it seems like a very small amount to aim for, especially with a R&P focusser. I suspect that this is the reason for dual speed Crayford focusers?

Captain Chaos

[Ambermile]

Sounds about right CC - the reason why the faster your scope the more critical the focussing. Also the reason why we (Artemis) spend so long making sure ccd's are 90 degrees to the light path. Over a (not particularly) big chip like the ICX285, any more than 50 microns from corner to corner difference will show up as out of focus across an image.

Arthur

[FLO]

Is this related to the diameter of optimum sharpness?

I understand that this 'sweet spot' varies according to how fast/slow a scope is but don't know how to calculate it.

[Kaptain Klevtsov]

Don't know about a "sweet spot". All I did was assume that the light coming in the front of the 'scope is 200mm diameter, it gets focussed to a point after 1000mm. This is because the 'scope has a 200mm aperture and at f/5, it has a focal length of (200 X 5) 1000mm.

As the light comes to a focus point, the cone shrinks as it goes away from the lens or mirror (same idea really) and finally ends up as a single tiny point at the focus position. I assume that a star should, ideally, illuminate a single pixel (if it hits it in the middle) so I worked out that the amount of movement away from the ideal position (either in or out) where the light spills onto adjacent pixels is 1/30 of a mm.

It turns out that this is true for all f/5 'scopes as its the f/ ratio that detemines the angle of the cone and all f/5 'scopes have the same cone angle. This is obvious really as the f/ ratio is the ratio of the length of the cone to the diameter of the base, if you follow my terminology.

The longer f/ ratio 'scopes will have more leeway in the focusing as the cone angle is less steep.

Now I need to figure out how a focal reducer affects things as the FR moves with the focuser. I can't imagine it making things easier though.

My f/5 'scope with the focal reducer will have a very blunt light cone, so that if the 'scope turns out to be f/2.5, I'll have 1/60th of a mm either side of focus to play with. Not a lot really, is it?

As Arthur points out, you need to have the camera CCD exactly at right angles to the 'scope axis within better than these tolerances or the focus will not be the same across the image.

Captain Chaos

[FLO]

The point of the cone as it enters the eyepiece has a diameter of optimum sharpness which varies according to the focal ratio of the scope (different to the exit pupil).

I don't fully understand what is happening or its significance .... ?

[Ambermile]

The focus point of the scope is where the EP goes Steve... once the light comes through to your eye it is parallel, not a cone anymore. That's what the EP does - in effect magnifying that point of focus.

Arthur

[FLO]

I understand that - an eyepiece straightens the cone and irons out abberations.

Its the diameter of useful sharpness (seperate to the exit-pupil) as it enters the ep that I am after - I think :scratch:

[Ambermile]

Ooo - that depends on the quality of the optics... as in how flat that focus point in the cone is. If you got bad optics you get coma, which is another way of saying that the image at the focus point in the cone is distorted to a greater or lesser extent. If you can imagine your image comes to a focus over an area of (say) 10 mm, then in an ideal world you can look at all of it. In bad systems you can only look at the middle. A barlow looks at the middle but also magnifies what's there, hence a darker image with (possibly) more aberation. A reducer does the opposite and tries to look at the whole image but compresses it over a smaller area, so a brighter image but more aberation as you can now see the nasty bits on the edge of the image.

Arthur

[FLO]

Thanks Arthur,

I know everything you say is true but its not quite what I am after.

Its my fault, I'll do some research and come back with a better worded question.

[FLO]

Rather than continue hijacking CC's thread (sorry), I have reposted the question here:

http://stargazerslounge.co.uk/index.php?board=24.0