Converting finite optics to infinity corrected?

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

enricosavazzi
Posts: 1474
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Converting finite optics to infinity corrected?

Post by enricosavazzi »

The following is (for now) just a thought exercise.

Finite optics (e.g. finite microscope objectives) are optimized to project an image at a fixed distance behind the objective (e.g. 160 mm). They usually don't work well if we substantially change this distance. First curvature of field usually appears, then other aberrations add to the problem.

Infinity-corrected optics are optimized to project an image on a focal plane located at infinite distance behind the objective. A tube lens (basically a convergent system) is added behind the objective to shorten this distance to a practical tube length (often 200 mm).

The purpose of this exercise is asking whether adding a divergent optical group behind a finite objective can turn it into an infinity corrected one. I am not sure about this, but the same divergent group should work with all objectives designed for the same tube length. Please comment if this is not true and every objective requires a different "negative tube lens". For example, what about a -160 (or -170) mm FL divergent group added behind finite 160 mm TL objectives?

An advantage of infinity optics that we often use is that by changing the FL of the tube lens we change the magnification of the system without pushing the objective outside its design specifications. With a negative tube lens followed by a positive one of different FL, at least in principle we should be able to do the same with finite objectives.

Incidentally, a focal length multiplier is basically a telescope with a convergent front group and a divergent rear one. A "speed booster" is a reversed telescope of the same type. We already use both to change magnification without bringing a lens outside its optimal magnification range. Would it make sense, or be advantageous in some way, to instead use custom optics like the ones discussed here? I don't think that many of us would like to mix finite and infinity objectives on the same microscope, but there might be more useful applications.

A logical extension of this concept would be applying it to high-resolution, highly optimized repro, copy and scanner lenses that allow only a narrow magnification interval before IQ starts deteriorating.

I am aware that finding a ready-made suitable negative tube lens can be so difficult as to make the idea impractical. There is however a chance that there are technological applications that use this type of imaging optics, in which case it might be possible to scavenge suitable optics from the second-hand market. A highly corrected convergent group is easy to find (tube lenses, Raynox add-on lenses and repurposed legacy telephoto lenses are examples), but where can we get an equally highly corrected divergent group? Parts of discarded camera lenses are unlikely to work, because they are usually designed to introduce heavy aberrations that are subsequently corrected by other elements, and would not work satisfactorily when separated from the rest of the lens.

PS - what about "telan" lenses? What about barlow lenses for telescope eyepieces?

Any comments?
--ES

Pau
Site Admin
Posts: 6051
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

what about "telan" lenses?
My Zeiss microscope uses a telan lens+tube lens system to create an infinite space to put intermediate optical elements like filters, beamsplitters or magnification changers without altering the 160 correction.

I've tested the opposite: infinite corrected objectives do work well removing the telan lens while maintaining the tube lens, so your idea will likely work well. Because that lenses AFAIK are only bundled with old microscopes they likely could limit the quality image circle. You could calculate the FL required and try doublet or triplet negative achromats of that FL
Pau

Ichthyophthirius
Posts: 1152
Joined: Thu Mar 07, 2013 5:24 am

Re: Converting finite optics to infinity corrected?

Post by Ichthyophthirius »

enricosavazzi wrote: An advantage of infinity optics that we often use is that by changing the FL of the tube lens we change the magnification of the system without pushing the objective outside its design specifications. With a negative tube lens followed by a positive one of different FL, at least in principle we should be able to do the same with finite objectives.
Hi Enrico,

Yes, a Zeiss Optovar magnification changer works like this: http://www.microscopy-uk.org.uk/mag/img ... 04.bmp.jpg

Negative bottom telan lens
Telescope turret
Positive to telan lens

The two telan lens systems actually seem to be achromats (two elements cemented together) in the case of Zeiss.

Leitz even build a zoom system for the Orthoplan www.ronaldschulte.nl/files/Variotubus-513-115.pdf

The Nikon trinocular heads for Optiphot etc. have a negative lens at the base of the dovetail that create an infinity space behind them. http://www.photomacrography.net/forum/v ... p?p=206784 This lens is also an achromat at best (might even be a single lens; I can't tell).

My understanding is that the negative lens system can be quite simple if you go far up the tube length. The closer to the objective, the better the system has to be.

Regards, Ichty

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

Re: Converting finite optics to infinity corrected?

Post by rjlittlefield »

Ichthyophthirius wrote:My understanding is that the negative lens system can be quite simple if you go far up the tube length. The closer to the objective, the better the system has to be.
This puzzles me. I would expect it to be the other way around: closer is simpler.

The reason is that close to objective, the lens needed to convert finite to infinite can be weaker and smaller diameter. For example to convert 160 mm finite to infinite, a negative lens located at objective only needs to be about -150 mm focal length and 12 mm diameter (matching the objective's rear aperture). But 100 mm higher, it will need to be -50 mm focal length, and almost as wide as the eyepiece field stop.

What have I missed?

--Rik

Ichthyophthirius
Posts: 1152
Joined: Thu Mar 07, 2013 5:24 am

Post by Ichthyophthirius »

Hi Rik,

I can't give you a reference for that. Just something I was told; the design requirements are supposed to be higher (lens form; surface quality etc.) the closer you get to a highly corrected system like a microscope objective. Further away, a suboptimal design was, supposedly, less likely to deteriorate image quality.

You don't think that's the case? It felt intuitive to me. But I'm happy to be corrected on this.

Regards, Ichty

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

Post by rjlittlefield »

Correct, in this case I do not think that closer is harder.

For intuition, mine says that if I were asked to look for two different lenses, one with FL=-150 and D=12, the other with FL=-50 and D=18, with equal levels of aberration, the long narrow one would be a lot easier to find and cheaper to buy. This is because the long narrow lens will have much shallower curves.

Diagrams like the one at http://www.microscopy-uk.org.uk/mag/img ... 04.bmp.jpg can become a bit of a trap for thinking about lens designs. This is because the only rays that are drawn are the ones for the central point in the field. As a result, they give the impression that lens elements must be wide at the objective and can get narrower toward the sensor. But when one also takes the edges of the field into account, then it becomes clear that lens elements near a typical objective can be quite narrow (the same diameter as the objective's rear aperture), while lens elements near the image may have to be significantly larger (approaching the diagonal of the sensor).

--Rik

soldevilla
Posts: 684
Joined: Thu Dec 16, 2010 2:49 pm
Location: Barcelona, more or less

Post by soldevilla »

I usually use finite lenses in front of photography lenses. The working distance is reduced a bit, but the quality I get is generally better than using the finite lens alone.
As barlow lenses, apart from the classic ones, there are the Televue optical systems and the photographic duplicators (very cheap those of M42) are systems of 4 to 6 perfectly apochromatic and aplastic lenses, at least in theory.

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

Post by rjlittlefield »

soldevilla wrote:I usually use finite lenses in front of photography lenses. The working distance is reduced a bit, but the quality I get is generally better than using the finite lens alone.
I am intrigued by this report. It sounds like you are dragging the objective somewhat away from its design point, but getting improved results. This behavior is not expected, if the objective designers did their job well.

I am wondering if there are some interesting tradeoffs involved, such as significantly improved corners at the cost only slightly degraded centers.

Can you clarify what your optical configuration is, and what "generally better" means?

--Rik

soldevilla
Posts: 684
Joined: Thu Dec 16, 2010 2:49 pm
Location: Barcelona, more or less

Post by soldevilla »

Yes, that is my perception. I believe that my objectives (a finite Nikon, probably not the most expensive in the store) have a small useful field and on the chip of my Canon 1100D they produce corners with a lot of chromatic and spherical aberration. The x20 is terrible. And placed in front of my Vivitar 100-200 the field is much more pleasant to see, although perhaps you are right and I lose some quality in the center.

My setup is very simple. The Vivitar attached to the Canon with a mount converter and a threaded adapter on the front of the telephoto lens to thread the microscope lenses there. Interestingly, I manufactured a coupling on my lathe for my non-microscope objectives, and they do not work at all, they only work alone, at a certain distance from the chip and with nothing between them.

Post Reply Previous topicNext topic