Where can one find a Coddington-like lens?

[jsmoove]

Im looking for a lens that focuses on its own plane, they were mostly used for Stanhopes back in the day to focus on pictures applied to the end of the lens.

https://en.wikipedia.org/wiki/Coddington_magnifier
https://en.wikipedia.org/wiki/Stanhope_(optical_bijou)

"This was a Plano-convex handheld lens whose focal length was the length of the overall lens, thereby greatly magnifying anything applied to its flat end"

Where does one find a modern lens that can accomplish the same thing?

 

[jsmoove]

I guess it is the "back focal length" that needs to be at a minimum to be able to focus as close as possible to the surface. Can someone confirm this ?

 

[AgX]

One answer of how to obtain such lens is at the end of that article...

 

[jsmoove]

Hey, sorry im not seeing where you're mentioning....from https://en.wikipedia.org/wiki/Coddington_magnifier ?
I dont see where they mention the back focal length or how to obtain one
Someone else mentioned to me that a glass sphere lens with a high refractive index over 2 would have an exceptionally short back focal length.
But spheres have distortion, whereas it appears the coddington does not because of its groove, but im not sure about its focal length.....either way, just need one that can magnify 10x like a loupe directly on its own plane without too much distortion and with a minimal 'back focal length'

 

[AgX]

But spheres have distortion, whereas it appears the coddington does not because of its groove,...

Distorsion is not caused by a lens being spheric. Also distosion is independent of size of aperture, thus a lens being aspheric will not solve the problem.
One thing to consider is the location of the aperture.

 

[jsmoove]

In what way for the aperture? I mean a complete sphere, a sphere ball
Im not understanding how you get a minimal back focal length, other than having a flat lens, one side flat and the other convex
I was looking at: https://www.edmundoptics.com/resources/application-notes/optics/understanding-ball-lenses/

 

[Nodda Duma]

From an optical shop as a custom order.

Coddington lenses are not commonly found in off-the-shelf catalogs like Edmund Optics, not unless some company has decided to start selling oddball items that sit on their shelves forever since the last time I was searching for a COTS lens at work (3 weeks ago).

You should spend a bit of time reading about and understanding the Thick Lens Equation, which you can then use to specify the lens you need. From there you can request a lens supplier to fabricate a custom optic to your specifications.

I’m assuming that the Coddington magnifiers available online don’t work for you, since sources for them start popping up as the second item in a google search for the term “Coddington lens”

 

[jsmoove]

My confusion is that im not sure now if a Coddington lens is what I need, I had thought that a coddington lens was flat on one side, but pictures on google show otherwise. And if the lens focuses on its plane, whether flat or not flat.
There also seems to be confusion for coddington vs stanhope, because not all stanhopes were coddington, yet they put a micro-picture on one end.
Im not finding answers on if this is the type of lens that focuses on its own plane without aberration.
I will check out the Thick Lens Equation.

 

[Nodda Duma]

No lens focuses without aberration. Particularly a simple lens like what you are searching for and even more particularly a lens which focuses on its own back surface (which is what I assume you meant by "on its own plane").

To clear up your confusion: Coddington is a type of lens. Stanhope is a type of optical device. Analogous to, for example, "achromat" and "telescope".


I'm saying this as a lens designer with many years' experience and who read your posts both here and on places like reddit: You either need to be looking at a ball lens, or if you don't like that, then select and bond a short focus Fresnel lens masked down to about f/5.6 or even f/11 and bonded to an acrylic block with the same length as the focal length of the Fresnel (accounting for the index of refraction of the acrylic block).

One side note: Much of your aberrations are due to field curvature. A curved focal plane will make the resulting image look better. This is easiest to do probably. Much of the rest of the aberrations will be due to spherical aberration (get an asphere) and color correction (achromatize). To correct these and basically step up from my suggestions above is $$$$$. That's 5 figures, to be clear, to design and fabricate one example, based on what you've described you want.

It can be done, just a matter of how bad you want it. In short: You can't get around physics or the real cost to design and fabricate custom optics.

Edmund optic 45-539 is probably your best bet, with a 4mm or 2mm image circle. This is the optical layout of that lens. Your eye is to the left. The colored lines are tracing the rays through the system to that 4mm (as modeled) diameter image circle

 

[Nodda Duma]

Note that you could encase all of the lens in epoxy except for the actual parts of the surface that the light rays pass through. So you could have a square ... thing ... with a bump for the front optical surface that you are looking through.

The simulated image of a 4mm tall picture on the back of the lens. This is what you would see looking through it.

If you want smaller, here's the simulated image of a 2mm tall picture on the back of the lens. This is what you would see looking through that.

 

[jsmoove]

Thankyou so much, i appreciate your expertise. This is extremely helpful! So it does have to be a sphere, how interesting, I never would have guessed that. All new territory for me.
So my question is then, is you had a larger diameter sphere, say an inch wide ($$?), can you still get a short back focal length? Does the diameter affect the image or is it just a matter of proper scaling?
Thanks for running your program, super cool to see what the image would look like.
I was looking at https://www.knightoptical.com/stock...lenses/ball-and-half-ball-lenses/ball-lenses/
The shortest back focal length is 0.05 (im assuming this is as close as you'd get?) mind you the diameter is only 0.6, so its tiny and probably for microscopic purposes I imagine.
The application im thinking of is to actually laser etch an image on the lens itself like a lithophane, its more of a data transmission test im curious about attempting with a mobile phone. So I figured that the lens would at least need to be a bit bigger just to accommodate the etched image itself, because im not entirely sure how small of a unit measurement you can get with laser etching. Its more that im trying to think of a single material solution, otherwise I would go the microphoto route and stick one on the back.
Also, I was looking at: which is 10x magnification, so if I can get something close to this with the sphere that would be dandy.
As for putting it in a cube, "with a bump for the front optical surface that you are looking through" sounds exactly what I want and what I was thinking. So when you say " except for the actual parts of the surface that the light rays pass through" is there some sort of formula for determining how much of the sphere could pop through the top of the cube?
So itd be a little cube with an image etched on the back basically that you could put up to your phone and retrieve the data.

 

[Nodda Duma]

Don’t over think it or over complicate it. I wrote exactly what you should be considering. K.I.S.S. Optics are complex enough.

The diameter of the ball lens that I show above is 1” diameter. The proportions are accurate so you can measure off that screenshot (just like old school off a map). It’s showing a 4mm tall image. If you want a smaller lens then scale down the image height as well.

Buy one and start tinkering. I could write 1000 words and you’d still know less than if you held one in your hand.

 

[jsmoove]

For sure....I dont see a 1"/25mm diameter ball with a short back focal length on edmund optics however. The Edmund optic 45-539 you mentioned is 1.5mm in diameter
I mean that I think I'd need more actual surface area if I'm going to etch an image into the ball not just paste one on the back like stanhopes. I was thinking at least an inch
Ill look around

 

[Nodda Duma]

Hmm. I guess I f’ed up reading the diameter on that listing. Probably had 5 different things going on. In any case, find a 1” diameter ball lens and you’ll get a similar result.

 

[jsmoove]

Ok cool, I wasn't clear on whether the diameter actually affected the image.
But one thing I am unclear about is when you say that the sphere can be put in epoxy or similar, would the image be on the back of the sphere or on the flat part of the epoxy, and does it make any difference? Is there more "image" in focus the bigger the sphere?
Also, im not sure how to calculate the actual magnification, to fit the etched image fully in the FOV when putting the sphere/cube up to a mobile phone lens.
I read https://microcosmos.foldscope.com/?p=33098 , so I understand that bigger sphere equals smaller magnification, so how to get 10x, im not sure.
I guess I can just try a bunch of different sizes, I just know that too small of a sphere may be difficult to work with

 

[jsmoove]

Just did this super quick in tinkercad: https://imgur.com/a/kAU1W6S
This is what you mean right?
Also perhaps sortof related but the opposite way around, a "sorotama lens" : https://soratama.org/en

 

[Nodda Duma]

Yep. Image goes on the spherical surface (read what I wrote about a curved image plane)

 

[jsmoove]

How does one achieve a curved focal plane in this instance?
Im reading: https://photographylife.com/what-is-field-curvature
Obviously the more image in focus the better....sorry if I'm not understanding correctly.
So I can etch an image on the back of a (some diameter) sphere, and plop it into either epoxy, glass, or acrylic with the top popping out and it should all work alright?
Still haven't found a ball lens that is over 10mm yet.

 

[Maris]

How about a "dome magnifier'. These things are put directly down on the paper to magnify fine print. They also make for nice paper weights. Glass and plastic versions exist.

 

[jsmoove]

@Maris Half dome seems applicable too since my understanding is that the lens has a short back focal length being flat....
Lost in the details a bit here, but more or less I just need a lens that focuses directly on the surface where my image will be, was hoping for 10x the magnification and a clear enough image that doesnt look like the magnifier on the left (unless correctable by software losslessly):
http://www.bvda.com/images/content/6-evidence-collection/k2000_v_k21000.jpg

 

[jsmoove]

@Nodda Duma Hey again, regarding the simulation you gave above for the spherical ball...was that for a diameter of 10mm? If I were to look into a 20mm diameter ball with an index of 2, could I get an image larger than 4mm on the back? I see the Edmund Optics 45-359 you recommended, but its 1.5mm diameter?

 

[Nodda Duma]

Without looking at the model, yeah I think so.

If it were me I’d get my hands on a practical diameter lens like 1” and then just do some testing to see what is feasible. Sometimes it’s faster and more appropriate to simply hands-on experiment.

 

[bdial]

The Edmund lens is 15mm in diameter, not 1.5, if that helps.
Surplus shed looks to have a number of Plano Convex choices that are similar, and between 15 and 25mm diameter;
lens finder

 

[jsmoove]

@bdial I think its this one though: https://www.edmundoptics.com/p/15mm-diameter-n-lasf9-ball-lens/6032/
Since I'm looking at ball lenses

 

[bdial]

Interesting, I googled the number and an Edmund 15mm PCX showed up.
These look like just the thing, except perhaps for the size. That they are on clearance might be a big plus.