star focus v moon focus

[iwols]

hi using autofocus which works very well on stars/galaxies,but after moving to the moon with same focus found moon slightly out of focus,would you expect this,just making sure my autofocus on stars/galaxies is working ok

[bobro]

The Moon is at the same infinity focus as stars/galaxies. Could it be the brightness of the Moon that makes it look as though focus is soft? Putting an aperture mask (can be a piece of cardboard with a hole) on the scope reduces brightness.

[AKB]

4 hours ago, bobro said:

The Moon is at the same infinity focus as stars/galaxies.

Is that really true?  The moon’s radius is ~1,000 miles, range 250,000 miles.  So middle to edge is a distance change of 1 in 250.   I’m sure it depends on your optical system, but this must be detectable?

[ollypenrice]

I have rarely found lunar and stellar focus to be identical.

Olly

[bobro]

1 hour ago, AKB said:

Is that really true?  The moon’s radius is ~1,000 miles, range 250,000 miles.  So middle to edge is a distance change of 1 in 250.   I’m sure it depends on your optical system, but this must be detectable?

 

35 minutes ago, ollypenrice said:

I have rarely found lunar and stellar focus to be identical.

I stand corrected (must look closer next time).

Thanks guys.

[billyharris72]

1 hour ago, AKB said:

Is that really true?  The moon’s radius is ~1,000 miles, range 250,000 miles.  So middle to edge is a distance change of 1 in 250.   I’m sure it depends on your optical system, but this must be detectable?

I'm not sure how this works to be honest. I'm pretty sure angular size has no effect on focus (or we could make the same argument about M33). The thing about having the rays coming in parallel is that they come in parallel across the mirror / lens from every point on the object, so the thing that would matter would be the size of the objective (more properly f ratio).

Please correct me if wrong on this though - it's been known!

I also reckon on this basis that focus should be identical, while agreeing with Olly that it never seems to work out that way. To what extent that's just an illusory result of seeing (and trying to focus to a tolerance that is less than seeing will allow) I have no idea.

Billy.

[kirkster501]

59 minutes ago, ollypenrice said:

I have rarely found lunar and stellar focus to be identical.

Olly

I think that's your tinkering nature Olly (just like me) ?  There can be no difference with objects that distant?  There's always that urge to try and get it better when pointed at the moon/Jupiter when in fact the focus achieved on the star was perfect.

[ollypenrice]

7 minutes ago, kirkster501 said:

I think that's your tinkering nature Olly (just like me) ?  There can be no difference with objects that distant?  There's always that urge to try and get it better when pointed at the moon/Jupiter when in fact the focus achieved on the star was perfect.

It may be but I never put what 'ought to be' ahead of what I find 'is.' Elevation may have something to do with it. It might be an effect in the eye/brain, too. Perhaps a camera would find no difference?

Olly

[Star101]

My findings

I find my C11 requires adjustment in the focus for the moon compared to the stars.  I have even found that I have needed to adjust focus for the center of the moon compared to the outer edges.

When I did a 14 panel mosaic of the moon, the center was in focus but the outer circumference was not! Now, maybe this was change in air temperature/pressure etc as I am in the UK. I would need to redo the process. But, personally for my setup, I do find stars and moon are not the same focus. They are close, but not same.

Just my findings.

[AKB]

59 minutes ago, billyharris72 said:

I'm pretty sure angular size has no effect on focus

Optical aberrations aside, it should not.  But range does, and there is a difference between 249 kmiles, 250 kmiles, and infinity...

34 minutes ago, Star101 said:

I find my C11 requires adjustment in the focus for the moon compared to the stars.  I have even found that I have needed to adjust focus for the center of the moon compared to the outer edges.

Indeed, so.

[Cosmic Geoff]

It should be possible to settle part of this argument by using optical formulae to calculate the shift of focus caused by the moon being 240,000 miles away rather than at infinity.

The focus may also shift with time as the scope cools down.

[AKB]

2 hours ago, Cosmic Geoff said:

It should be possible to settle part of this argument by using optical formulae to calculate the shift of focus caused by the moon being 240,000 miles away rather than at infinity.

A wise suggestion!

The (thin) lens equation:

1 / focal_length = 1 / image_distance + 1 / object_distance

If we take focal length as 1 metre, for example, and object distance as very much greater than 1, then this is approximately:

image_distance = 1 + 1 / object_distance

If object distance is infinity, then the image distance equals the focal length.

For the moon, object distance = 400,000 km = 4 x 10^8 m.

image_distance = 1 + 2.5 x 10^-9 m

so the image plane shifts by 2.5 nanometres.

I don't think that even the fastest optical system has a focusing requirement of that!

So, mea culpa, although the image distance IS sensitive to object range, the effect is small (unless, of course, some of the above is wrong.)

Apologies.

Tony

_______________

PS: However, there may be something in real-world systems which is not captured by the above approximation ?

[billyharris72]

4 hours ago, Cosmic Geoff said:

It should be possible to settle part of this argument by using optical formulae to calculate the shift of focus caused by the moon being 240,000 miles away rather than at infinity.

Funnily enough, I've been thinking along similar lines (Bank Holidays, eh?) and taken a slightly different approach based on using the hyperfocal distance theorem, which, with a suitably tight circle of confusison, ought to give a reasonable indication of the point beyond which more accurate focusing has no payoff.

I've assumed a (totally arbitrary but reasonably sensitive) instrument of F=1000mm, N =4 (where N is the f ratio). Shooting on an ASI1600, with a x10 crop and printing the result at 16 x 20 inches (so about as hard as it gets outside of exhibition work).

Gauss suggested that for an 8 x 10 print viewed at standard reading distance (much more able to show blur than a non-4K monitor) the appropriate circle of confusion would be the sensor (ok, film) diagonal / 1,750. That's the point at whicha human of normal visual acuity would be unable to detect a difference at normal reading distance. This is widely accepted as unrealistically precise (1500 is common for critical work in digital media, but lets stick with 1750). That gives a very challenging CoC in mm of 21.9/10/3,500 = 0.0006.

Plugging those (unpalatably challenging numbers into the hyprerfocal formula H=(f^2/(Nc))+f gives a result of 399.5km.

In all seriousness, with a 250mm x 1000mm scope, 400km is infinity from an optical perspective. In real terms, we could get away with much less most of the time.

Billy.

[Alien 13]

I think the difference noted might be down to how focus is obtained, when imaging the Moon we have regions of high contrast that can effectively be used to make sure the focus is spot on without any doubts. Focusing a Star is usually via software tools that look at the light curve and make a best guess or a bat mask which again is a case of judgement if the pattern isn't perfect, I know that if I use liveview zoomed in on the very faintest star there is only one focus position and this is for my kit is exactly the same as as I get with focus on the Moon.

Alan

[michael8554]

Obviously the bottom of a moon crater is further away than the top of a moon Mountain, so moon focus is never certain........ ?

Seriously, I agree with Alan's explanation.

Michael 

[bobro]

4 hours ago, Alien 13 said:

I think the difference noted might be down to how focus is obtained, when imaging the Moon we have regions of high contrast that can effectively be used to make sure the focus is spot on without any doubts. Focusing a Star is usually via software tools that look at the light curve and make a best guess or a bat mask which again is a case of judgement if the pattern isn't perfect,

Could it be the other way round? E.g. when focus is obtained using a star it means focusing on a specific point on the imaging plane, which isn't too difficult using a mask or software. However, when attempting to focus the Moon, it is a larger object that is spread across the imaging plane. With a real and imperfect scope there is probably (depends on equipment) no single point of perfect focus for an object that spans the imaging plane to some degree (perhaps due to optics or imaging plane alignment) requiring a visual judgement of perfect focus that is perfect across the plane.

 

[Alien 13]

Just now, bobro said:

Could it be the other way round? E.g. when focus is obtained using a star it means focusing on a specific point on the imaging plane, which isn't too difficult using a mask or software. However, when attempting to focus the Moon, it is a larger object that is spread across the imaging plane. With a real and imperfect scope there is probably (depends on equipment) no single point of perfect focus for an object that spans the imaging plane to some degree (perhaps due to optics or imaging plane alignment) requiring a visual judgement of perfect focus that is perfect across the plane.

 

I agree with this too, I think what it shows is that you need to understand your own camera/scope and dont just assume that if you have perfect focus on the Moon that the Stars will be 100% too and vice versa.

Alan

[alan4908]

Well on my very limited imaging with the Moon......but a little more with DSO's - I've found that the moon will be in focus provided that you use a stellar source and you autofocus with care. What I've found is:

1. It is best to choose a focus star near the zenith - this is to reduce the impact of seeing. Seeing introduces uncertainty into the autofocus operation and so you need to try to minimize this. The larger the air mass, the more uncertainty you will create. 

2. The focus star needs to be selected such that it is exposed by your camera to be on the linear part of the your camera's response - if you pick overexposed stars you can get misleading focus results. 

3. I also found that some form of automated focus convergence is very useful to minimize the effect of seeing. 

I found that the free version of FocusMax performs the focusing very well, however, I use ACP to find the focus star. 

Alan