52mm Tubes, and other adapters.

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Is that graph for finite lenses or infinite lenses? And how can you have a >200mm change in tube length without degrading the image if the nominal tube length is 200mm???

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Lou Jost wrote:Is that graph for finite lenses or infinite lenses? And how can you have a >200mm change in tube length without degrading the image if the nominal tube length is 200mm???
Agree, maybe it's just a total range of 200mm around the 200mm tube length? So +-100mm around 200mm, maybe? And what is IQ degrading criteria? 1/root2 times something?

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

I suggested, that you can't equate:

change in tube length (mm) on a finite objective arrangement,
with
tube lens distance (mm) from the sensor, in an infinite arrangement.

In both cases you're expecting the objective to focus "somewhere wrong", but they're not the same situation. Rather than attempting an analysis beyond pencil sketch and a couple of calculations which give me confidence to go that far, I'll leave it there.

They are fairly similar though, and surely, in both cases, higher NA's will be non-linearly more critical?

The much quoted graph is for finite tube lengths, from times before infinite lenses were used, I think. Rik has commented further on it, but the reference is hard to find.

I remember in a focus-by wire exercise, Rik's observation that you can't push a 50x NA 0.55 as far as you can a 10x NA 0.25, before the IQ falls away.

Qualitatively that's something I can confirm.
However, putting a NA 0.8 objective on a 200mm tele and finding the focus not at infinity on the tele, made me repeat a stack. I couldn't tell the difference, though.
I just checked the tele - I'd guess it was 1-2mm out.


In a pm, one member is using a 210mm 40x NA 0.5 , on the front of a 200mm tele lens and says it's fine. So wahddoIknow!? :lol:
Chris R

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Post by mjkzz »

I think that for an infinite objective, the light ray coming out of it are "parallel" rays as if they come from far far away.

The job of a tube lens is to converge this "parallel" rays from infinity and project them onto its focal plane as a sharp image.

So I think it is important to place the tube lens at its focal length away from the sensor to have a sharp image projected on sensor.

Based on above, if tube lens focal length is shorter but it is still placed at its focal length away from sensor, the image projected should be sharp albeit smaller (less magnification)

However, not sure why, I have accidentally left tube lens placement about 2mm closer to sensor with my 50x 0.45NA objective, and I got somewhat OK image and I was puzzled.

Maybe an infinite lens has some other rays coming out of it and somehow get converged by the tube lens, but these are not the rays the objective is designed for and thus less IQ.

Maybe someone can do a comparison test -- same subject, same lighting, but one with 5mm off and one with dead on and see what happens.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Chris, Lou, Peter,

So it seems one can wander away from the ideal 200mm for inf corrected objectives (Mittys) more so with the lower NA types. How much so is probably not well defined, and best observed by experiment.

I do know that the 5X Mitty works well with the Raynox 250 (125mm) @ 125mm, but I haven't pulled the Raynox away from 125mm though to see how the IQ behaves.

Mike

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Mike there is almost no problem with using different focal length tube lenses as long as you use the corresponding tube length, as you did; as you say, that just affects coverage (though there is some internet literature suggesting longer is better). The deviation that I have been worried about is the deviation of the tube length from the focal length of the tube lens. A while ago I did some experiments with a 10x 0.28 Mitutoyo and found that major deviations of this kind did matter, but a millimeter or two didn't.

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Post by mjkzz »

Mike,

I think as long as the tube lens used can project "parallel" rays from infinity (coming out of an infinite objective) onto the sensor placed at its focal length away, ideally, it should not matter, though in practice, every piece of equipment has its limits.

The issue/question/puzzle, for me at least, is when the sensor is NOT placed at tube lens' focal plane, even though you might get a sharp image, I suspect that the image is formed by rays not designed/intended for, ie, not by "parallel" rays, therefore, IQ might suffer from this deviation.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Yes, it's not a binary "OK"/"Not OK" situation. It isn't easy to compare, either. when you alter the sensor to tube lens distance, focus changes, so you have to run two stacks and compare, unless you can look at a tiny part of the image which is all in focus in one shot.
You're interested in the change in the optimisation of the setup, not the focus, which is dealt with by itself/separately.

It's a bit like trying to adjust a correction collar on an objective. You can't just "make it sharp" because focus changes at the same time as optimisation.
Chris R

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

mjkzz wrote:Maybe an infinite lens has some other rays coming out of it and somehow get converged by the tube lens, but these are not the rays the objective is designed for and thus less IQ.
mjkzz wrote:The issue/question/puzzle, for me at least, is when the sensor is NOT placed at tube lens' focal plane, even though you might get a sharp image, I suspect that the image is formed by rays not designed/intended for, ie, not by "parallel" rays, therefore, IQ might suffer from this deviation.
Peter, I think your puzzlement is that you're too focused (pardon the pun) on the idea that the output of the objective has to be parallel rays. The objective is just a lens. It happens to be optimized for a particular focal arrangement. For an infinity objective, that optimum has parallel rays coming out the back. But like any other lens, if you move the subject a little bit closer to the objective, then the rays coming out the back will diverge a little from where they were, and if you move the subject a little bit farther from the lens, the rays coming out the back will converge a little. If you extend the tube lens less than its focal length, then in order to form a focused image, the subject has to move farther away from the objective so that the rays coming out back of the objective are already converging enough that the tube lens on its shorter extension can finish the job. Nothing dramatic happens, it's just that you've moved the objective away from its designed optimum point, so some aberrations are no longer corrected as well as they are at the optimum.

With objectives, the first aberration of interest is spherical (SA). The loss of correction goes as the fourth power of NA. High mag objectives have larger NA, and that's why they are more sensitive to being dragged away from their designed focus points.

--Rik

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Post by mjkzz »

Thanks Rik,
It happens to be optimized for a particular focal arrangement. For an infinity objective, that optimum has parallel rays coming out the back.
OK, this confirms the two "maybe" of mine in your quotes that if tube lens is not placed at right place, rays forming the image are not the optimum rays, thus not what the objective is designed for even though it works in practice.
With objectives, the first aberration of interest is spherical (SA). The loss of correction goes as the fourth power of NA. High mag objectives have larger NA, and that's why they are more sensitive to being dragged away from their designed focus points.
If the rays forming the image are not what the objective is designed to handle, would this 4th power of NA rule still apply? Could it be worse or better?

I do not know much about optics, but all of these highly specialized ingredients and formulas that APO or Fluror lenses utilize, they got to be formulated around its design point. So I guess a lens design must be very sensitive to its usage conditions.

So my point is, though I do not know optics, in general when a system is designed and optimized around a set of constraints, would any of the theory established within the constraints still apply when there is (are) deviation(s)? if not, how sensitive? better or worse?

I mean, people pay multiple times of money to get, maybe, last 5% improvement of performance and these deviations might just invalidate that little improvement. Just a thought.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Sure
Expect > understand > test & measure > refine understanding

then decide how much the result matters to you. If it matters a lot, it'll probably be expensive :lol:
Chris R

Post Reply Previous topicNext topic