making wide angle lens for ulf camera

[toguko]

I would like to make a wide angle aplanat (Rapid Rectilinear) lens that could cover 20"x60" negative size (image circle = 1800mm including the front movements). The lens has to be wide angle so that the vertical view could be normal (because of the panoramic ratio of the negative). According to my calculations, the best focal length would be 1000mm with the angle of view = 84°. I would like to ask the experts in optics on this forum that could advice me how to start. I would highly appreciate any help.

And this the prototype of the camera I am building:
114769

[Mark Sawyer]

Just so you know, Wide Angle Rectilinears weren't Rapid Rectilinears. WAR's were usually f/16, while RR's were usually f/8. Both used achromatic doublets front and rear, but if you're using black and white film, you can probably get away with single elements front and rear.

[VPooler]

www.surplusshed.com has the materials needed

[Lachlan 717]

I'd really reconsider front movements, particularly tilt/swing. These necessitate a larger image circle than rear movements, and will be a considerable PITA when focusing.

[toguko]

Just so you know, Wide Angle Rectilinears weren't Rapid Rectilinears. WAR's were usually f/16, while RR's were usually f/8. Both used achromatic doublets front and rear, but if you're using black and white film, you can probably get away with single elements front and rear.

I will be using paper negatives hence Periskop lens would be too slow. It is my second choice though.


www.surplusshed.com has the materials needed

Perhaps they do have the materials but first I need to know which lenses I need.


I'd really reconsider front movements, particularly tilt/swing. These necessitate a larger image circle than rear movements, and will be a considerable PITA when focusing.

I will be using only rise and fall front movements.

[ruilourosa]

Hello

I have the same problem... i would love to have a rapid rectilinear made out of two achomats with around 1000mm but i have found some problems...

surplusshed has NOT any achromats of around 2000mm with a viable diameter (i would like at least 80mm or a bit more)

i have tried to find on other online optic stores but it´s hard... i didn´t find in any...

i turned myself to the low power (low diopters) achromat close up lenses and i have succeded, i made a +- 620mm out of two minolta achromat close ups, but 620mm it´s not a 1000 (i know i could make a 1000mm out of a +1 element alone, building a chevalier achromat, but i need more quality, at least a bit more, and also a bit more speed...), i would love to find two identical, coated 2000mm achromats with 80mmm (f:12) or ideally with 125 mm (f:8)

my 620 has a perfectly acceptable quality but it was not cheap to make... (around $100)

low quality achromats may be used but i have no knowledge of low power lenses (less than +1)

let me know if you know something :)

Cheers


Rui Lourosa

[Mark Sawyer]

I will be using paper negatives hence Periskop lens would be too slow. It is my second choice though.


A periscopic lens would be the same speed as an equivalent lens made with Achromats, you'd just lose the achromatic corrections. These would be unnoticeable in monochrome; Kodak once used achromatic meniscus lenses in its folding cameras, but switched to single lenses when they found there was no difference in the results.

[Amedeus]

Surplus Shed doesn't have a steady supply of large diameter (>125mm) lenses of any type. Once and a while you get lucky on a special to find out the focal length is rather short.


www.surplusshed.com has the materials needed

[Amedeus]

What is your budget ?

Why do you believe 1000mm with an angle of view of 84 degrees to be best ? Is this limited by bellows draw ? Wide angle lenses are harder to make ...


I would like to make a wide angle aplanat (Rapid Rectilinear) lens that could cover 20"x60" negative size (image circle = 1800mm including the front movements). The lens has to be wide angle so that the vertical view could be normal (because of the panoramic ratio of the negative). According to my calculations, the best focal length would be 1000mm with the angle of view = 84°. I would like to ask the experts in optics on this forum that could advice me how to start. I would highly appreciate any help.

And this the prototype of the camera I am building:
114769

[alexn]

What lachlan said is my instant thought... walk around the camera, apply front tilt, walk around the camera, check ground glass, adjust focus slightly, walk back around the camera, adjust front tilt a little more, walk back, check.. throw arms in the air, walk back to the front of the camera, zero movements, stop down to f/295, shoot and deal with the diffraction as opposed to the frustration.. Rear movements would be a much better option for something of this scale I would have thought..

[Tim Meisburger]

I think he said only rise and fall, so tilt is irrelevant. Mark, if I understand you correctly, there is no need to use achromats if you are shooting black and white? So the dispersion of different colours will have no affect in monochrome? I don't really understand optics, but I do know I can get cheap +.5 diopter meniscus lenses here from optometrists. I really want to shoot the moon...

By the way Tom, that is a hell of a camera, and I'm looking forward to seeing your results!

[Lachlan 717]

Just pin-hole it....

[Dan Fromm]

The dispersion of different colours will have no affect in monochrome?

Of course chromatic aberration has an effect in black and white. Some soft-focus lenses (Berthiot Nebulor, Boyer Opale and, IIRC, Puyo-Pullgny) use chromatic aberration to get the effect and are completely unusable with color materials.

[Tim Meisburger]

Thanks Dan. Don't want to hijack the thread, so will continue this elsewhere.

[Mark Sawyer]

Unless you're designing the lens to be flat-field, it's likely to have a curved focus plane, making front movements less effective. And it's likely to have a sharper center "sweet spot", which front movements will move around. And with marginal coverage, you may vignette when taking the lens off axis with front movements. So yeah, design for rear movements...

[hoffner]

So yeah, design for rear movements...

Mark, please reread the post n.5. As Toguko only wants the rise and fall movements it is only logical to build it on the front standard instead of moving the rear beast. In any case, the image result is the same.

[Lachlan 717]

Mark, please reread the post n.5. As Toguko only wants the rise and fall movements it is only logical to build it on the front standard instead of moving the rear beast. In any case, the image result is the same.

Actually, the image result is not the same. Front rise/fall changes the lens' position relative to the composition, thus changing the image. Rear rise/fall simply uses a different section of the original projected image without changing the composition.

Whether the resulting differences are of consequence enough to worry about is up to the Shooter, but it is incorrect to state the images are the same.

[hoffner]

While you're right in your theoretical assessment I was naturally speaking in view of the OP's particular camera. From the picture of the beast I suppose that it will be used in a landscape mode only and for landscape photography. The image difference between the front and rear rise and fall will then be practically negligible while the difference in the construction of the camera is overwhelming.

[toguko]

Hello

I have the same problem... i would love to have a rapid rectilinear made out of two achomats with around 1000mm but i have found some problems...

surplusshed has NOT any achromats of around 2000mm with a viable diameter (i would like at least 80mm or a bit more)

i have tried to find on other online optic stores but it´s hard... i didn´t find in any...

i turned myself to the low power (low diopters) achromat close up lenses and i have succeded, i made a +- 620mm out of two minolta achromat close ups, but 620mm it´s not a 1000 (i know i could make a 1000mm out of a +1 element alone, building a chevalier achromat, but i need more quality, at least a bit more, and also a bit more speed...), i would love to find two identical, coated 2000mm achromats with 80mmm (f:12) or ideally with 125 mm (f:8)

my 620 has a perfectly acceptable quality but it was not cheap to make... (around $100)

low quality achromats may be used but i have no knowledge of low power lenses (less than +1)

let me know if you know something :)

Cheers


Rui Lourosa


does the diameter of the lens only determine the f stop? I will definitely let you know if I find anything!



A periscopic lens would be the same speed as an equivalent lens made with Achromats, you'd just lose the achromatic corrections. These would be unnoticeable in monochrome; Kodak once used achromatic meniscus lenses in its folding cameras, but switched to single lenses when they found there was no difference in the results.


From what I read ('Photographic Optics' by Monckhoven), the periscopes need small diaphragms and the useful aperture is around f64, otherwise 'the field is much less flat and the image much less sharp'. Also there is a small inconvenience of correcting the focusing every time due to the periscope's chemical focus. The aplanats could be used almost wide open giving sharp image even at the edges. I would be really happy if I could use f22 every time.



What is your budget ?

Why do you believe 1000mm with an angle of view of 84 degrees to be best ? Is this limited by bellows draw ? Wide angle lenses are harder to make ...


At the moment my budget is $1000 but obviously I would like to spend as little as possible as I have lots of other spendings for this project (material for bellows, huge glass, three tripods, lots of wood, special paper, chemicals, some kind of backpack/sledge for hiking and probably thousand other little things). If there is no other way, I could probably organize more though.
The reason I need a wide lens is because of the panoramic ratio of the negative (20"x60"). I will be shooting landscapes so if I want to have vertical view to be almost normal, the horizontal has to be (very) wide. So for this size, according to my calculations, focal length of 1000mm and angle of view of 84 degrees should be perfect. Obviously I am open to some reasonable changes though.



What lachlan said is my instant thought... walk around the camera, apply front tilt, walk around the camera, check ground glass, adjust focus slightly, walk back around the camera, adjust front tilt a little more, walk back, check.. throw arms in the air, walk back to the front of the camera, zero movements, stop down to f/295, shoot and deal with the diffraction as opposed to the frustration.. Rear movements would be a much better option for something of this scale I would have thought..


It would be impossible to do any back movements with this size of glass on, too heavy. Also I need front rise&fall to adjust the perspective - it might be difficult on many occasions to move the whole camera as high as needed. I don't need tilt&shift.



I think he said only rise and fall, so tilt is irrelevant. Mark, if I understand you correctly, there is no need to use achromats if you are shooting black and white? So the dispersion of different colours will have no affect in monochrome? I don't really understand optics, but I do know I can get cheap +.5 diopter meniscus lenses here from optometrists. I really want to shoot the moon...

By the way Tom, that is a hell of a camera, and I'm looking forward to seeing your results!


Could you please tell me where I could find these +0.5 diopter meniscus lenses? (just in case) What max diameter do they have available?



Just pin-hole it....


I'm planning to use paper negatives. With pinhole, the exposure would be too long which would result in overexposing the sky and underexposing the subject. Also I need much better quality and have more control on the final image.



Unless you're designing the lens to be flat-field, it's likely to have a curved focus plane, making front movements less effective. And it's likely to have a sharper center "sweet spot", which front movements will move around. And with marginal coverage, you may vignette when taking the lens off axis with front movements. So yeah, design for rear movements...


As I mentioned before, rear movements are impossible and not needed.



Mark, please reread the post n.5. As Toguko only wants the rise and fall movements it is only logical to build it on the front standard instead of moving the rear beast. In any case, the image result is the same.


thank you :)



Actually, the image result is not the same. Front rise/fall changes the lens' position relative to the composition, thus changing the image. Rear rise/fall simply uses a different section of the original projected image without changing the composition.

Whether the resulting differences are of consequence enough to worry about is up to the Shooter, but it is incorrect to state the images are the same.


And that's why I need front rise&fall.




Can anyone please enlighten me what does the angle of view depend on? Is it the thickness of the lens, radius or the space between the lenses? Thank you

[wombat2go]

Hi, Toguko,
I suggest you download the freeware "OpticalRayTracer" It runs on java, so will work on any computer o/s.
It is primitive, but has all the main functionality for ray tracing, and does not require much learning to get it working.
To model a variable stop, I just use a thin element with RI=1 and change its radius.

You could set up a model with your camera dimensions and try different lens scenarios.
Last year Dan Fromm pointed out links that have prescriptions for the Periscop etc.
I was able make models of those prescriptions on OpticalRayTracer rather easily, and see the abberations.

[Dan Fromm]

Actually, the image result is not the same. Front rise/fall changes the lens' position relative to the composition, thus changing the image. Rear rise/fall simply uses a different section of the original projected image without changing the composition.

Whether the resulting differences are of consequence enough to worry about is up to the Shooter, but it is incorrect to state the images are the same.

Please explain further. I ask because as I see it shifting the lens' axis so that it is no longer centered on the film plane (moving the front standard) is equivalent to shifting the film plane so that the lens' axis is no longer centered on it (moving the rear standard). What am I missing?

[LF_rookie_to_be]

(...) special paper, chemicals, (...)
I'm planning to use paper negatives.

Just out of curiosity, which type of paper will you be using? Are you going to contact-print the paper negatives?

[brucetaylor]

I know little about optics, but if the OP is looking for large + .5, 1, etc. close up lenses for this project, 138mm diameter close up diopters are a standard size for motion picture cameras. Fairly easy to find used too.

[Lachlan 717]

Please explain further. I ask because as I see it shifting the lens' axis so that it is no longer centered on the film plane (moving the front standard) is equivalent to shifting the film plane so that the lens' axis is no longer centered on it (moving the rear standard). What am I missing?

Dan,

In moving the lens up/down (or side to side if using shift) from an original point, you change where the image is being formed. This will, therefore, change the relationship between near/far objects. To see this phenomenon, place your fingers in front of your eyes, in line, one behind the other so that the front finger blocks the back one. Then move them, in unison, up and/or down and you will see the previously blocked finger appear.

This does not happen with rear movements as you are just choosing a different section of the static projected image. In other words, the lens has not changed its position relative to the subject.

[Dan Fromm]

Lachlan, the lens projects a cone of rays. When it covers a circle larger than the one that circumscribes the film, decentering movements without unacceptable loss of image quality in the film's corners are possible.

With that in mind, I'll ask you again. What's the difference between moving the cone of rays the lens projects across the film using decentering movements of the front standard and moving the film's position in the cone's base using decentering movements of the rear standard?

Yeah, sure, front decentering movements displace the lens' axis so that it no longer points at exactly the same point in the subject and rear decentering movements don't but the difference between the effects at the film plane is nil except at quite near distances. This is why either front rise or rear fall can be used to remove inconvenient foregrounds from the frame.

And just for you I did the experiment. 6" lens with ample coverage for 2x3, 2x3 Cambo, subject (back fence with plants in front of it) ~ 40' away. Full right shift at the rear, front centered. Full left shift at the front, rear centered. Same view on the GG. What did I do wrong or misinterpret?

[Maris Rusis]

The easiest way to get a wide angle effect on a view camera is to add a wide angle conversion lens to the front of a normal lens. These afocal lenses are abundantly available (on eBay for example) in a wide variety of sizes and compression ratios. The downside includes compromised (more or less) image quality and the possibility of some barrel distortion. Stopping down the prime lens seriously improves image quality which may easily be enough when enlargements are not envisioned.

[Mark Sawyer]

Of course chromatic aberration has an effect in black and white. Some soft-focus lenses (Berthiot Nebulor, Boyer Opale and, IIRC, Puyo-Pullgny) use chromatic aberration to get the effect and are completely unusable with color materials.

The chromatic aberration from these lenses will be far less pronounced than the spherical aberration. As I said, Kodak gave up on producing more expensive doublets when they found people couldn't tell the difference in black and white. And since the OP said he was going to shoot paper negatives, which will be orthochromatic, effect will be minimized even further.

[toguko]

Hi, Toguko,
I suggest you download the freeware "OpticalRayTracer" It runs on java, so will work on any computer o/s.
It is primitive, but has all the main functionality for ray tracing, and does not require much learning to get it working.
To model a variable stop, I just use a thin element with RI=1 and change its radius.

You could set up a model with your camera dimensions and try different lens scenarios.
Last year Dan Fromm pointed out links that have prescriptions for the Periscop etc.
I was able make models of those prescriptions on OpticalRayTracer rather easily, and see the abberations.


I do have this program already. Any chance you could send me these prescriptions or just a print screen from one of your designs, just to give me something I could start from, please?



Just out of curiosity, which type of paper will you be using? Are you going to contact-print the paper negatives?


I still need to find paper that would be good for making calotypes. If making my own paper negatives is too hard for this size, I will get Slavich single fiber Unibrom. I'm planning to do salt prints or contact prints if I choose Slavich.



I know little about optics, but if the OP is looking for large + .5, 1, etc. close up lenses for this project, 138mm diameter close up diopters are a standard size for motion picture cameras. Fairly easy to find used too.


thank you very much for this information!



The easiest way to get a wide angle effect on a view camera is to add a wide angle conversion lens to the front of a normal lens. These afocal lenses are abundantly available (on eBay for example) in a wide variety of sizes and compression ratios. The downside includes compromised (more or less) image quality and the possibility of some barrel distortion. Stopping down the prime lens seriously improves image quality which may easily be enough when enlargements are not envisioned.

Unfortunately I need to control the barrel distortions as much as possible, otherwise I would have used single lens. Also it is not easy and cheap to get normal lens for this image circle.



The chromatic aberration from these lenses will be far less pronounced than the spherical aberration. As I said, Kodak gave up on producing more expensive doublets when they found people couldn't tell the difference in black and white. And since the OP said he was going to shoot paper negatives, which will be orthochromatic, effect will be minimized even further.


Would using f22 stop on periscope lens give you reasonably sharp and flat image?




The diameter determines the f stop?
The diopter determines the focal length?
What about the thickness, sphere radius, lens radius, curvature and space between the lenses?

[Lachlan 717]

Lachlan, the lens projects a cone of rays. When it covers a circle larger than the one that circumscribes the film, decentering movements without unacceptable loss of image quality in the film's corners are possible.

With that in mind, I'll ask you again. What's the difference between moving the cone of rays the lens projects across the film using decentering movements of the front standard and moving the film's position in the cone's base using decentering movements of the rear standard?

Yeah, sure, front decentering movements displace the lens' axis so that it no longer points at exactly the same point in the subject and rear decentering movements don't but the difference between the effects at the film plane is nil except at quite near distances. This is why either front rise or rear fall can be used to remove inconvenient foregrounds from the frame.

And just for you I did the experiment. 6" lens with ample coverage for 2x3, 2x3 Cambo, subject (back fence with plants in front of it) ~ 40' away. Full right shift at the rear, front centered. Full left shift at the front, rear centered. Same view on the GG. What did I do wrong or misinterpret?

You are changing the near/far relationship when you raise/lower/shift the lens. It projects a different image. Just as you see a different view standing up compared to when you're sitting down.

It has little to do with the projected cone; it has everything to do with what the lens "sees" and how the components within the image are aligned relative to each other.

[toguko]

Dan Fromm and Lachlan: gentleman, please move your discussion somewhere else.

So if I get two 138mm +0.5 close up filters and put them opposite to each other, will I get a periscope lens with the focal length of 1000mm? What would be the angle of view?

[ruilourosa]

your angle will depend on the spacing you do between the lenses, less spacing more curvature of field, more spacing less angle coverage... but those 138mm are expensive... prepare to pay around 200€ for a less than good brand and more for schneider, schneider even makes achromats in this size (for around 1300£€)

if you use 77mm close up filters you will get a f:12,5, that you can stop down a bit,

why don´t you try to find a copy of monckhoven optics treaty

and ray tracers are great if you are about to choose a type of glass and a radius and a lens type and ask someone to cut your elements to size or order (expensive as hell)

if you are doing experiments with things you already have (and don´t know what kind of glass is it for instance or the curviness of the surfaces) you are better with a totally empirical process, experiment with optics until you get decent results

one advice: star always with identical focal lenght´s from both elements and identical designs, assimetrical designs are more difficult to work (less contrast) and introduce distortion

cheers

rui

[wombat2go]

Hi, Toguko,
For your request post #28, here is an OpticalRayTracer .ini file for a Periscop of focal length 88mm.
I am unable to locate the original prescription on my files here , perhaps you could search Dan's posts here or apug.

Don't hold me to it, but I think you can scale it to any focal length by multiplying all dimensions including spacing, element radii and spherical radii by (new_focal_length/88).
leave the RI and Abbe unchanged.
Then tune it for your camera design, using the diagonal of your film holder.

I am not sure if that will yield a practical design.
Anyway I expect you will have to work backwards making a model with glass that you can find available.

https://app.box.com/s/9urbhau9gu7xwr2nzc0i

[Dan Fromm]

meandering marsupial, Eric Beltrando's collection of prescriptions can be found at www.dioptrique.info

toguko, thread drift is a fact of life. Posters who start discussions don't own them and can't control them. Learn to live with it.

[Amedeus]

So if I get two 138mm +0.5 close up filters and put them opposite to each other, will I get a periscope lens with the focal length of 1000mm? What would be the angle of view?

You will get 1000mm focal length at roughly f7.5




What about the thickness, sphere radius, lens radius, curvature and space between the lenses?


SInce you plan working with off-the-shelf lenses, you don't have control over the above (and more) factors and other than the space between the lenses. This means you don't have much control over the inevitable aberrations. The latter will define what your acceptable angle of view is ... all depends on your expected visual quality criteria. Illumination doesn't mean acceptable "coverage". DOn't expect much more than 45 degrees with these simple designs and off the shelf glass unless you have access to more exotic varieties.

If you're set on a wide angle approach, then find a working design from the www.dioptrique.info and scale all dimensions as wombat2go suggested. Challenge still is to then find the right glass combinations.

From my working experience, periscope and rectiinear are easiest to play with from a two element perspective.

[wombat2go]

Some searching indicates this company
http://www.newportglass.com/achromat.htm
http://www.newportglass.com/share.htm
(I have no affiliation)
provides semi finished concave lenses for amateurs in the form of achromat kits with grinding tools, in the 4 to 6 inch dia f/10 range.

Most internet info on amateur lens grinding is concentrated on making plano-concave mirrors.

Anybody here with experience in lens grinding?

[Lachlan 717]

Dan Fromm and Lachlan: gentleman, please move your discussion somewhere else.

So if I get two 138mm +0.5 close up filters and put them opposite to each other, will I get a periscope lens with the focal length of 1000mm? What would be the angle of view?

With any due respect, mind your own business. If the OP wants us to move this somewhere else, then let him/her tell us. If you don't like what we're writing, report us and let the Moderators for THEIR job.

[hoffner]

With any due respect, mind your own business. If the OP wants us to move this somewhere else, then let him/her tell us. If you don't like what we're writing, report us and let the Moderators for THEIR job.

Lachlan,
please reread the post n.1 - Foguko is the OP.

[Nathan Potter]

A good source that I have used for specialty lenses is Edmund Optics USA. Singlets and achromats coated or uncoated can be had at reasonable prices, although 2000 mm. FL and 80 mm dia. might be extreme. I'd experiment using much smaller pairs (and a lot less expensive), then scale to the dimensions required paying attention to the tradeoff between element spacing and coverage.

I think Edmund will still custom make the full size pairs for you at as low a cost as anyone.

It sounds as though you're after a double convex pair in Periskop form or for extreme wide angle a primitive Goerz Hypergon.

Nate Potter, Austin TX.

[toguko]

your angle will depend on the spacing you do between the lenses, less spacing more curvature of field, more spacing less angle coverage... but those 138mm are expensive... prepare to pay around 200€ for a less than good brand and more for schneider, schneider even makes achromats in this size (for around 1300£€)

if you use 77mm close up filters you will get a f:12,5, that you can stop down a bit,

why don´t you try to find a copy of monckhoven optics treaty

and ray tracers are great if you are about to choose a type of glass and a radius and a lens type and ask someone to cut your elements to size or order (expensive as hell)

if you are doing experiments with things you already have (and don´t know what kind of glass is it for instance or the curviness of the surfaces) you are better with a totally empirical process, experiment with optics until you get decent results

one advice: star always with identical focal lenght´s from both elements and identical designs, assimetrical designs are more difficult to work (less contrast) and introduce distortion

cheers

rui


yes, they are expensive indeed. thank you for your suggestions.



Hi, Toguko,
For your request post #28, here is an OpticalRayTracer .ini file for a Periscop of focal length 88mm.
I am unable to locate the original prescription on my files here , perhaps you could search Dan's posts here or apug.

Don't hold me to it, but I think you can scale it to any focal length by multiplying all dimensions including spacing, element radii and spherical radii by (new_focal_length/88).
leave the RI and Abbe unchanged.
Then tune it for your camera design, using the diagonal of your film holder.

I am not sure if that will yield a practical design.
Anyway I expect you will have to work backwards making a model with glass that you can find available.

https://app.box.com/s/9urbhau9gu7xwr2nzc0i


thank you very much! that's very helpful!



meandering marsupial, Eric Beltrando's collection of prescriptions can be found at www.dioptrique.info

toguko, thread drift is a fact of life. Posters who start discussions don't own them and can't control them. Learn to live with it.


yes, I know this website. It's really good!



You will get 1000mm focal length at roughly f7.5



SInce you plan working with off-the-shelf lenses, you don't have control over the above (and more) factors and other than the space between the lenses. This means you don't have much control over the inevitable aberrations. The latter will define what your acceptable angle of view is ... all depends on your expected visual quality criteria. Illumination doesn't mean acceptable "coverage". DOn't expect much more than 45 degrees with these simple designs and off the shelf glass unless you have access to more exotic varieties.

If you're set on a wide angle approach, then find a working design from the www.dioptrique.info and scale all dimensions as wombat2go suggested. Challenge still is to then find the right glass combinations.

From my working experience, periscope and rectiinear are easiest to play with from a two element perspective.


That's what I thought as well that the angle of view wouldn't be too wide. +0.5 close up filter is the least curved one.



Some searching indicates this company
http://www.newportglass.com/achromat.htm
http://www.newportglass.com/share.htm
(I have no affiliation)
provides semi finished concave lenses for amateurs in the form of achromat kits with grinding tools, in the 4 to 6 inch dia f/10 range.

Most internet info on amateur lens grinding is concentrated on making plano-concave mirrors.

Anybody here with experience in lens grinding?


that's actually really interesting. thanks for this link.



A good source that I have used for specialty lenses is Edmund Optics USA. Singlets and achromats coated or uncoated can be had at reasonable prices, although 2000 mm. FL and 80 mm dia. might be extreme. I'd experiment using much smaller pairs (and a lot less expensive), then scale to the dimensions required paying attention to the tradeoff between element spacing and coverage.

I think Edmund will still custom make the full size pairs for you at as low a cost as anyone.

It sounds as though you're after a double convex pair in Periskop form or for extreme wide angle a primitive Goerz Hypergon.

Nate Potter, Austin TX.


I did contact them. no reply.


---


Lachlan, Dan: I'm sure a lot of people is interested in the difference between front and back camera movements. It would be such a waste to bury such valuable information inside a topic about the diy wide angle lens for ulf camera.

[pierre506]

Just found that someone made a Hypergon / Globe type lens.
The lens information is f22 / 250mm / 134°，image circle 1000mm.
Hoping it's not a joke.

115022

[ruilourosa]

Hey!!!!!! more information please!!! i want to make one!!!! just need the lenses!!!!!!!!!!! :) (i think this could be a serious business, cutting some lens elements to make simple and historical lenses like aplanats, periskops, hypergons, triplets... even if the tolerances aren´t that good... i think they would sell!!!!!)

cheers

rui

[MDR]

Aplanats, hypergons and triplets are far from simple the might only have 2 to 4 elements but they require tight tolerances. The Hypergon has to be made by hand as the curvature is too extreme for most machines. A periskop is doable so is a wide angle meniscus, Thomas Suttons liquid filled wide angle lens might be worth some investigation (http://www.mhs.ox.ac.uk/sphaera/index.htm?issue8/articl7) but it has the same problems as the Hypergon.
Here's a link to a Fuji Patent for a two element WA lens https://depatisnet.dpma.de/DepatisNet/depatisnet?action=pdf&docid=US000005067803A
A Goerz Patent for a two element lens : https://depatisnet.dpma.de/DepatisNet/depatisnet?window=1&space=menu&content=treffer&action=pdf&docid=AT000000099601B&Cl=3&Bi=1&Ab=&De=1&Dr=3&Pts=&Pa=&We=&Sr=&Eam=&Cor=&Aa=&so=asc&sf=vn&firstdoc=2&NrFaxPages=3&pdfpage=1

[grzybu]

Maybe you could bend the negative? ;)
Then it will be much easier to fix the field curvature problem because you have panoramic format.
Ground glass could be made of acrylic so it should be possible to bend it a little.
Just the idea.

[Dan Fromm]

Bending the film to fit a cylinder, easy. Bending the film to fit a sphere, much harder.

[grzybu]

But in this case film width is few times greater than height, so horizontal field curvature will be much stronger.

[ruilourosa]

I´m not planning to open a lens business!!!!! i just want to play with some focal distances and field coverage, with some images in focus at least... tight tolerances are for carl zeiss or schneider!

[Dan Fromm]

But in this case film width is few times greater than height, so horizontal field curvature will be much stronger.

20" x 60" if the OP's fantasy is realized. Not enough for your idea to be useful.

[LF_rookie_to_be]

Here's a 1900mm double achromat, available and brand new:

http://www.edmundoptics.eu/optics/optical-lenses/achromatic-lenses/large-precision-achromatic-lenses/54569

Just wondering, what approximate image circle does this lens cast? And do you simply stop it with a hole cut in a piece of black cardboard?

[pierre506]

Here's a 1900mm double achromat, available and brand new:

http://www.edmundoptics.eu/optics/optical-lenses/achromatic-lenses/large-precision-achromatic-lenses/54569

Just wondering, what approximate image circle does this lens cast? And do you simply stop it with a hole cut in a piece of black cardboard?

The problem is how big your camera is and how long your bellows is.
I got a 2700mm F12 double achromat lens, too. The diameter is over 220mm. It must need a room to swallow the lens.

[LF_rookie_to_be]

I got a 2700mm F12 double achromat lens, too. The diameter is over 220mm. It must need a room to swallow the lens.

Yes, but does the image circle cover the room's wall? :)

[Fr. Mark]

http://www.re-inventedphotoequip.com/Prices.html

fairly reasonable ULF lens prices, likely pretty soft at edges, max focal length is I think 790 mm which'd be pretty wide for the OP's design.

There's a book with a title like primitive photography which describes in some detail how to make fairly simple lens assemblies i.e. a cooke triplet, where to put the f-stop that sort of thing, but it assumes you buy lenses from some sort of maker of lenses.

When I get impatient to try a new format i.e. 5x8 inches I build a pinhole camera.

As to making lenses with sloppy tolerances, I think you are wishing for something the universe is not obligated to give you. You are asking for a violation of the rules of basic physics/optics. It is hard enough to make half decent lenses when you are willing to get things aligned straight/co-linear etc and the lenses ground right.