Close-up lens construction question & confusion

xkaes · Mar 22, 2024

I know that most close-up lenses are a simple, one-piece, concave, meniscus lens -- of varying "strength" -- and that others have more than one element to control for various optical aberrations.

But I've run into a set of Ketnor close-up lenses that have me confused. They are simple, one-element lenses, but instead of being concave, they are pretty flat -- even though they create the appropriate close-up effect.

Here are two photos comparing the weakest #1 CU lens (on the left) -- first the front, then the rear -- to a Tiffen #1 (on the right). The Tiffen has the obvious concave shape, while the Ketnor does not. They both reflect TWO images of the single light bulb indicating ONE glass element.

The Tiffen shows the light bulb as expected, on both sides, but the Ketnor reflection suggests a nearly flat surface.

Can anyone explain the differences, optically? Is one better than the other???

reddesert · Mar 22, 2024

A simple thin lens has a focal length given by

1/f = (n - 1) * (1/a + 1/b)

where f is focal length, n is index of refraction, and a and b are the radii of curvature of the two sides. For a concave surface (like one side of a meniscus lens), the radius of curvature is negative. You can assume n=1.5 for the crown glass used in a close-up lens.

So for example, a +1 close-up lens has a focal length of 1000 mm. One could make this as: a biconvex lens with each surface with radius of curvature 1000mm; a plano-convex lens with one surface flat and the other with radius 500mm; a meniscus lens with radii of curvature 250mm and -500mm, and so on.

It is well known that for box cameras with a simple lens, using a meniscus lens with the concave side towards the subject (larger distance, convex side towards the film at smaller distance) reduces some off-axis aberrations. I suspect that with a close-up lens, it's intended to be used with a relatively close subject and the main lens "focused" at some larger distance, so a meniscus with the convex side toward the subject is likely to be preferred, but all close-up lenses are a compromise.

Mark J · Mar 22, 2024

A meniscus landscape lens is not very relevant to this specific case. it uses a spacing from the stop, then a bending of the lens around the stop, to trade spherical aberration for improved astigmatism & coma in the field.
However when most close-up lenses are put on the front of an existing lens, the stop is still inside and there's no advantageous bending of the close up lens in that way, they are actually bent the opposite way. I guess the philosophy of the meniscus ones is to optimise the shape for a central object point, in which case spherical aberration will be minimised.

I haven't modelled this case of meniscus vs. plano-vex close up lenses, but given the weak power of them , the difference is likely to be small, especially when stopped down. Having a meniscus one with the concave facing towards the subject will probably be better if the used aperture is large and the field angle of the lens is low. However, if you put this eg. on a standard lens, with a decent field, then I wouldn't be surprised if the Ketnor one gave a slightly better result over the picture.

Without doing the analysis, this is my best guess.

xkaes · Mar 22, 2024

I appreciate these comments, but I can't quite fathom how the Tiffen #1 has a strong curvature (concave on one side and a similar convex on the other) while the Ketnor #1 is basically flat on both sides. The reflections in the photos demonstrate this.

I understand how the curvature on the Tiffen bends the light -- just like all the other CU filters I've seen, but the Ketnor lacks this curvature -- yet they both magnify to the same extent.

Mark J · Mar 22, 2024

The power of the lens depends upon the difference in curvature between the two sides. What you have here is two lenses with the same overall power, but different bending.

ic-racer · Mar 22, 2024

How do they perform? Is one better than the other?
Personally I use the Nikon close up lenses. Those have a concave/convex design whereby they are mounted in a ring twice as thick as most other close up lenses. I can't say if this makes them better or not.

xkaes · Mar 22, 2024

ic-racer said:
Personally I use the Nikon close up lenses. Those have a concave/convex design whereby they are mounted in a ring twice as thick as most other close up lenses. I can't say if this makes them better or not.

It does make a difference. Nikon's CU lenses are are achromatics -- two lenses cemented together -- one concave, one convex.

The Tiffen and the Ketnor that I'm talking about are simple, one-element, meniscus lens -- such as the CU lens in this diagram (marked #1.

The Tiffen #1 is exactly like this -- convex on one side and concave on the other. The Ketnor #1 is basically flat on both sides -- as confirmed by the reflections in the photos.

How can a nearly flat CU lens create the same magnification as a curved lens?

Mark J · Mar 22, 2024

xkaes said:
How can a nearly flat CU lens create the same magnification as a curved lens?

Because it's the same focal length.
I'll have to think of a way of explaining this better tomorrow.
Let me sleep on it.

ic-racer · Mar 22, 2024

xkaes said:
It does make a difference. Nikon's CU lenses are are achromatics -- two lenses cemented together -- one concave, one convex.

The Tiffen and the Ketnor that I'm talking about are simple, one-element, meniscus lens -- such as the CU lens in this diagram (marked #1.

View attachment 366225

The Tiffen #1 is exactly like this -- convex on one side and concave on the other. The Ketnor #1 is basically flat on both sides -- as confirmed by the reflections in the photos.

How can a nearly flat CU lens create the same magnification as a curved lens?

I use #1 and #2 which are not apochromatics but still have a thick mounting ring. Maybe they just made the #1 and #2 in a thick housing to match the "T" close up lenses?

Mr Bill · Mar 23, 2024

xkaes said:
I understand how the curvature on the Tiffen bends the light -- just like all the other CU filters I've seen, but the Ketnor lacks this curvature -- yet they both magnify to the same extent.

When you say "understand..." what do you mean, roughly? For example do you mean that you understand, more or less, how to do ray-tracing through the curved surfaces? Or perhaps you just mean that you can see visible curvature, and know that this can "bend" light (as opposed to a flat window, for example)? I'm guessing more like the latter.

Using reddesert's radius calculations for a +1 diopter lens (1,000 mm focal length) I calculated (roughly) how this would affect the thickness of a lens. If the closeup lens has a diameter of about 50 mm (~2 inches) then the center would need to be about 0.625 mm thicker than the edge. That's all; about the thickness of 2 or 3 business cards, or 0.025 inch. So on casual observation the non-meniscus +1 diopter lens is gonna seem pretty close to flat. Especially if this extra thickness is split with half on each side of the lens.

Mark J said:
The power of the lens depends upon the difference in curvature between the two sides. What you have here is two lenses with the same overall power, but different bending.

This is largely just a different way of saying that the center of the lens has a different thickness than the edges.

When Mark J (who apparently knows something about lens design) refers to "bending," think of it like this: imagine that you have a 50mm diameter lens that is made of a rubbery sort of material. Such that if you hold the edges with your fingers and press against the center you could distort it into the shape of a sunglass lens. The "power" of the lens remains the same but the shape can be changed. You wouldn't actually do this with a physical lens, but in the design stage such "bending" essentially defines different curvatures for the two surfaces.

Now, I know just enough about this sort of thing to be maybe a little bit dangerous, but as a general rule one wants the work, the redirection (by refraction) of light rays, to be shared more or less equally between lens surfaces. Said differently, to avoid any one surface doing excessive amounts of work. So I would guess that the issue of using a bi-convex, etc., vs meniscus shape for a closeup lens is related to this, perhaps related to the specific camera lens. Just a guess.

dynachrome · Mar 23, 2024

Apart from the scientific explanations of how close-up lenses work, there is the intangible factor of how a particular close-up lens works with a particular lens. Some lenses, like older Sigma zooms, came with matching close-up lenses. Early zooms did not, for the most part, focus very close. I have an old 82mm close-up lens which was made by Minolta for its very old 80-160mm zoom. I don't have that lens but the close-up lens fits my Konica Hexanon 400mm f/5.6 UC lens. It gets the 400 close enough for a tight portrait. I have a modest collection of close-up lenses. It includes both single element examples and multi-element ones. The multi-element ones are from Konica, Minolta, Canon, Nikon, Pentax, Topcon, Bronica, Olympus and maybe others. The larger size ones are mostly for getting longer lenses and zooms to focus closer. In a way, close-up lenses are like teleconverters. You have to try them with different lenses to see how they work. One of my favorite close focusing zooms is the Vivitar 100-200 f/4 with the matched teleconverter. With a close-up lens, it gets even closer. Years ago Modern Photography tested a number of macro lenses and also tested a 50/2 Nikkor with Nikon single element close-up lenses (0, 1, 2). The 50/2 with the single element Nikon close-up lenses was nearly as sharp as some of the macro lenses. Even single element close-up lenses have their place.

Mark J · Mar 23, 2024

I like the rubber lens analogy, Mr Bill.

xkaes · Mar 23, 2024

I've figured it out -- as can be seen in the diagram, A simple, single-element CU lens (left side) will have substantial curvature -- like my Tiffen. Much more than 2-3 business cards. An achromatic, two-element CU lens will be much flatter, on the rear, due to the additional element (right side). I had assumed that the Kelnor was a single element lens due to it's low price.

Mark J · Mar 23, 2024

Well... maybe, maybe not.
If it's only a 1 Diopter lens then it's doubtful it's achromatised . Is it a 1 D ?
It could just be like your left-hand lens above, but flattened out , so one side is plano.

If it feels like it's substantially thicker and heavier, then yes, it might be a doublet.

Dan Fromm · Mar 23, 2024

Mark J said:
If it feels like it's substantially thicker and heavier, then yes, it might be a doublet.

Counting reflections would answer that question.

xkaes · Mar 23, 2024

Mark J said:
Well... maybe, maybe not.
If it's only a 1 Diopter lens then it's doubtful it's achromatised . Is it a 1 D ?
It could just be like your left-hand lens above, but flattened out , so one side is plano.

If it feels like it's substantially thicker and heavier, then yes, it might be a doublet.

Minolta's No. 0 (#0.94) is an achomatic, two-element lens. So is the Leica Elpro 4 (#0.75) The diopter is unrelated to the number of elements in the optical design.

Mark J · Mar 23, 2024

No, I don't quite agree that it's got 'nothing to do with it' . There is more likelihood that a stronger close-up lens like a 2 D or 3D will have two or even three elements.
There's less need to achromatise a weak lens like a 1000mm focal length.

What is the diopter of the Kelnor lens ?

What is the diopter of the Minolta lens ?

ps. a plano/convex single lens is cheaper to make than a meniscus single lens like the one on the left above.

xkaes · Mar 23, 2024

As I mentioned, the Kelnor is a #1 and the Minolta is a #0.94.

While it's true that higher diopter CU lens benefit more from an achromatic design, there are plenty of lower powered CU lenses that use it. The CU lens for the Minolta APO 100-500mm zoom has a diopter of #0.33.

Mark J · Mar 23, 2024

Ah sorry, it was the '#' that threw me.
Yes, I can imagine Minolta might have put more effort into a CU lens for an Apo objective.

reddesert · Mar 23, 2024

I doubt that a +1 closeup lens from Ketnor, a generic brand that few have ever heard of, is an achromat. The better Nikon closeup lenses (called 3T through 6T) were achromats and noticeably thicker and more expensive than a generic. I think the Ketnor is just a plano-convex lens and that's why it looks flatter.

Whether a particular close-up lens works well with a particular primary lens is likely only measurable through user testing.

JPD · Mar 23, 2024

reddesert said:
I think the Ketnor is just a plano-convex lens and that's why it looks flatter.

Yes, in simple terms: if the back surface is plane then the front convex surface would be flatter too for a +1 lens compared to one with a concave back surface.

Maybe the flat surface makes the lenses cheaper to manufacture.

Focomatter · Mar 24, 2024

Being an underwater photographer I have used a number of close-up lenses of both the simple single lens type and achromats. They are used with some frequency in this field. Nikon has made both types - I have some of both. I have seen just one of the flat ones which I have been led to understand are older. Someone more senior to me (I am near 70) suggested that they were more common some time ago (as in many decades).

Close-up lens construction question & confusion

Privacy & Transparency

Privacy & Transparency