Reversing the Zeiss 50mm f/1.6 Photolithography Lens - Optics Math Questions

[Doppler9000]

There are specifications posted on an eBay listing for other copies of this lens, which is 1/10x.

Object to image distance is listed as 602mm, flange focal distance is 24.5mm, barrel length is 160.5mm, image field diameter is 14.5mm.

Would it be correct that reversing the lens to use at 10x that:

Object to image distance remains unchanged at 602mm

New working distance = old flange focal distance = 24.5mm

New flange focal distance
= object to image distance - old flange focal distance - barrel length
= 602mm - 24.5mm -160.5mm
=417mm

New image field diameter = image circle = 14.5mm x 10 = 145mm

The eBay listing has a fairly comprehensive set of specifications.

[FotoChris]

-Forget what I said-

[Doppler9000]

Well yes, mostly, if you reversed the lens/optical path you get the reverse of the magnification while the distances remain the same.

A few things to consider:
1. what's the working distance when reversing the lens? The distance is not measured from the surface from the last glass element but - as the name implies - the flange. So the real working distance is quite a bit shorter.

Are you referring to this lens in particular, or lenses in general? For this lens, the rearmost areas of the rear element appear to be very close to co-planar with the very back of the lens.

[Lou Jost]

That's right, Doppler. Everything just gets switched around. The lens will work fine. Here is an old thread I made about using a 5x Nikon Engineering photolithography lens in reverse:
http://www.photomacrography.net/forum/v ... =8&t=35669

[FotoChris]

-Forget what I said-

[Lou Jost]

Doppler, don't over-think it. You know the distance between image and object, that is the most important thing. Move the lens around until you get the best image, leaving the camera and subject in their correct places. Don't try to calculate anything, these are strange lenses and assumptions underlying common formulae will not apply.

[rjlittlefield]

Also most lenses (unless they're perfectly symmetrical) have a different focal length when used in reverse and when used for closeups/macro.

For example: When you take a 50mm enlarger lens and use it in reverse, you end up with a focal length of around 38mm and as a consequence the theoretical flange focal distance also changes.

Something is wrong here.

For a lens with the same medium on both sides, the focal length is also the same on both sides. Simply reversing the lens will not change the focal length.

Of course the principal points are usually located different distances from the two ends of the lens, so reversing the lens usually will change the focal distances.

Personally I like to use this calculator to determine the focal length: https://www.omnicalculator.com/other/focal-length

Um, that calculator appears to be based entirely on the thin lens formula, 1/S1 + 1/S2 = 1/f.

It essentially assumes that front and rear principal points coincide with each other.

That's the simplest possible assumption, suitable for getting you in the ballpark but seldom very accurate. And if it gives you the idea that simply reversing a lens changes its focal length, then that aspect is just plain wrong.

Don't try to calculate anything, these are strange lenses and assumptions underlying common formulae will not apply.

Yea verily! (I agree completely.)

--Rik

[FotoChris]

-Forget what I said-

[Pau]

As Rik says reversing a lens doesn't change its focal length (nor its mass or other physical properties)
Also it doesn't change when focusing it closer only by extension.
Focal length changes if you add an auxiliary lens (teleconverter, diopter...) or when you change focus by rotating the focusing ring of a lens with floating elements, but in both cases you're altering the lens formula so in fact they become different lenses.

How are you measuring it? (...and what are you actually measuring?)

[FotoChris]

-Forget what I said-

[Lou Jost]

Yes, we all know that focal length changes when you move the optical components around, as in modern macro lenses with internal focusing. But this is not the case we are talking about here.

[FotoChris]

-Forget what I said-

[Lou Jost]

Yeah but that's the same principle and it does not matter whether it's a modern lens or not.

It is NOT the same principal. Re-arranging elements is radically different from just reversing a lens. Zoom lenses change their focal length by re-arranging their elements; no one here would argue that those do not change their focal lengths.

Some modern lenses are just better corrected to reduce the amount of focus breathing BY moving around internal elements.

That's news to me. Edit: You are right about this! See below.

For me this was especially noticeable on the old Mamiya 645 50mm 4.0 shift

Please tell us did you come to this conclusion.

You're not just reversing the lens. You're also changing the distance between the lens and the sensor AND you're changing the distance between the lens and the object.

In this case we are interchanging the image and object. Nothing more.

[rjlittlefield]

How are you measuring it? (...and what are you actually measuring?)

by using the calculator at a known magnification and object distance.

This is one place the dispute arises: FotoChris is trusting the calculator to produce an accurate answer and it does not.

The reason it does not is that, for the sake of simplicity, the calculator assumes that the lens acts "thin".

An accurate calculator would have to use the "thick lens" model, which has two additional parameters called "principal points" from which focus distances are measured.

But the locations of principal points are seldom known, so it is traditional for calculators to just ignore them.

This trades off less accuracy for more convenience. For practical purposes, the simplification is usually a good thing. But if not recognized, it can also lead to some confusion.

...check the photonstophotos.net site and look at the changes in focal length when focussing.
A few examples:
The Sony 90mm 2.8 will reduce its focal length to 45.88mm at 1:1, which is a pretty substantial reduction.
The Canon 100mm 2.8 L IS USM will reduce its focal length to 74mm at 1:1, which is pretty good.
The Canon MP-E 65mm 2.8 goes down to 36.72mm at 5:1, that too is pretty good considering the magnification!

These are all true, and they result from moving lens elements.

And even (or rather especially) when there's no movement of internal less elements and the lens is "only" being extended it will reduce its focal length which is very noticeable when you pay attention - it's what people call "focus breathing" or "lens breathing".
https://www.youtube.com/watch?v=43asl1yYw2E

This is well scrambled so let me try to unpack it.

In standard optics terminology the term "focal length" describes a specific property of how the lens bends light. That property is not changed by reversing a lens or by changing its extension.

But in the photography community, those same words "focal length" are often repurposed to indicate the scale of an image that the lens produces.

In the video, this is what the presenter is doing when he equates focus breathing with changes in focal length.

Ironically, to avoid focus breathing requires a lens that does change focal length -- at least when the words "focal length" are used as in optics.

Here at photomacrography.net, the tradition is to use "focal length" in its optics sense -- the one that is not changed by reversing or extension.

--Rik

[FotoChris]

It's all fine by me. Really, I'm not here to commit heresy or ruffle feathers - and if that's the way it is, then that's fine.

I'm just telling you what works for me - not just in theory but for practical purposes with reproducible results.
If I were to remove a scanner lens I would probably just measure it at the magnification it was used - and then use that as the working focal length - but I'll leave it at that.

Maybe I'll just skip the Technical Discussion page in future, doesn't seem like I'm cut out for that. Probably best to take pictures instead