Lens Construction

[CMoore]

OK.....I have read a bunch on Wiki, and The Internet in general.
I know the answer is right in front of me, but I cannot find it in All The Text that pops up from a search.

With two of the "same" lens..... For example a Canon 35mm/f1.8 and a 35mm/f3.5.
Lens A = f1.8
Lens B = f3.5
What limits Lens B from being able to open up to f1.8.?

I am kind of wondering...why don't "they" just make all lenses capable of a 2.0 aperture.? Why are/is there frequently the "same" lens manufactured at different speeds.?
Thank You

 

[summicron1]

with a wide lens, look at how far from the edge of the lens the glass has to capture the light, bend it and send it on its way to the film, all while not introducing errors.

With a small opening -- f 3.5 or even less -- it is pretty easy to make a lens that is real sharp. There are some really long but slow telephoto lenses, non-zoom, that are only one 2-piece element of really spectacularly ground glass out at the front. Or consider a kodak folding camera that has a wide-open aprature of f-8. It does a pretty good job with a two-element lens.

But at f 2 you have a lot of work to get that light onto the film and still project a flat image (so the corners stay sharp) with no halo, coma or other errors, and all in focus -- remember, the center of the lens is closer to the film than the edges, which means the lens has to project the light different distances while still keeping the final image sharp, which is why lens construction gets pretty complex as you go wide, with exotic glasses that have unique indexes of refraction and so on. To design a lens in the old days took a lot of mathematical work by rooms full of mathematicians.

Modern 21mm lenses on a 35mm camera that are f 2 or even wider were impossible before computers to do the math came along.

 

[Gerald C Koch]

The following describes how the max aperture of a lens is calculated.

http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/stop.html
https://en.wikipedia.org/wiki/F-number
https://en.wikipedia.org/wiki/Aperture

 

[CMoore]

with a wide lens, look at how far from the edge of the lens the glass has to capture the light, bend it and send it on its way to the film, all while not introducing errors.

With a small opening -- f 3.5 or even less -- it is pretty easy to make a lens that is real sharp. There are some really long but slow telephoto lenses, non-zoom, that are only one 2-piece element of really spectacularly ground glass out at the front. Or consider a kodak folding camera that has a wide-open aprature of f-8. It does a pretty good job with a two-element lens.

But at f 2 you have a lot of work to get that light onto the film and still project a flat image (so the corners stay sharp) with no halo, coma or other errors, and all in focus -- remember, the center of the lens is closer to the film than the edges, which means the lens has to project the light different distances while still keeping the final image sharp, which is why lens construction gets pretty complex as you go wide, with exotic glasses that have unique indexes of refraction and so on. To design a lens in the old days took a lot of mathematical work by rooms full of mathematicians.

Modern 21mm lenses on a 35mm camera that are f 2 or even wider were impossible before computers to do the math came along.

Ahhhhhh.....OK. The tip of the iceberg.
I got so much info when I searched "Lens Construction", that it was/is kind of over-whelming.
But I get the gist of what you are saying.
Very interesting indeed. Another good reason/example of why it is better to have a "crappy" camera and a good lens, than the other way around.
I realize that lens building is a complex art, but when you see two lenses that look the same.......
Thanks again everybody.
I Appreciate It

 

[Xmas]

with a wide lens, look at how far from the edge of the lens the glass has to capture the light, bend it and send it on its way to the film, all while not introducing errors.

With a small opening -- f 3.5 or even less -- it is pretty easy to make a lens that is real sharp. There are some really long but slow telephoto lenses, non-zoom, that are only one 2-piece element of really spectacularly ground glass out at the front. Or consider a kodak folding camera that has a wide-open aprature of f-8. It does a pretty good job with a two-element lens.

But at f 2 you have a lot of work to get that light onto the film and still project a flat image (so the corners stay sharp) with no halo, coma or other errors, and all in focus -- remember, the center of the lens is closer to the film than the edges, which means the lens has to project the light different distances while still keeping the final image sharp, which is why lens construction gets pretty complex as you go wide, with exotic glasses that have unique indexes of refraction and so on. To design a lens in the old days took a lot of mathematical work by rooms full of mathematicians.

Modern 21mm lenses on a 35mm camera that are f 2 or even wider were impossible before computers to do the math came along.

No modern high speed lenses are dependent on three things

High refractive index and different dispersion glasses
Glasses with low melting points that are fluid close to melting points and can tolerate crush in ceramic pressing
High precision CNC grinding machines

 

[Xmas]

OK.....I have read a bunch on Wiki, and The Internet in general.
I know the answer is right in front of me, but I cannot find it in All The Text that pops up from a search.

With two of the "same" lens..... For example a Canon 35mm/f1.8 and a 35mm/f3.5.
Lens A = f1.8
Lens B = f3.5
What limits Lens B from being able to open up to f1.8.?

I am kind of wondering...why don't "they" just make all lenses capable of a 2.0 aperture.? Why are/is there frequently the "same" lens manufactured at different speeds.?
Thank You

Same reason a formula 1 car is faster than a town car cost more to make.

 

[Xmas]

I wonder which one is more difficult to make--a good lens or a man-made diamond. And I don't even know if there's such a thing as a man-made diamond.

Yes you can make diamonds, but deBeers still operate a cartel in gem stones.

https://en.m.wikipedia.org/wiki/List_of_synthetic_diamond_manufacturers

I like easier questions BTW.

 

[Nodda Duma]

No modern high speed lenses are dependent on three things

High refractive index and different dispersion glasses
Glasses with low melting points that are fluid close to melting points and can tolerate crush in ceramic pressing
High precision CNC grinding machines

The first of the three you list is the more important for design though I wouldn't quite call it a modern improvement. Computer-aided design was significantly more important for improving optical performance. Coming up behind that is the change to digital with imagers smaller than 35 mm film, allowing even better correction over the smaller image area. That stuff all makes my job easier than it used to be.

The latter two you list are improvements in manufacturability, only indirectly helping design.

 

[Nodda Duma]

I wonder which one is more difficult to make--a good lens or a man-made diamond. And I don't even know if there's such a thing as a man-made diamond.

Optics are easier: they were made long before diamonds.

Know what's even tougher? An optical diamond window. The largest one I've seen is 1" diameter and cost more than a small house to fabricate. Very interesting object. Very high thermal conductivity.

 

[bdial]

Optics are easier: they were made long before diamonds.

Know what's even tougher? An optical diamond window. The largest one I've seen is 1" diameter and cost more than a small house to fabricate. Very interesting object. Very high thermal conductivity.

That's an interesting tangent. What was it used for?

 

[Xmas]

The first of the three you list is the more important for design though I wouldn't quite call it a modern improvement. Computer-aided design was significantly more important for improving optical performance. Coming up behind that is the change to digital with imagers smaller than 35 mm film, allowing even better correction over the smaller image area. That stuff all makes my job easier than it used to be.

The latter two you list are improvements in manufacturability, only indirectly helping design.

My definition of modern includes the statement lenses Zeiss made pre WWII but did not productionise.
I did not say design I include sales volume per the OP query, which includes cost of Cosina and Leica formed aspheric high volume lenses.

I should have also included the multi coating too as a 4th enabler for the high refractive glass.

Cheap in retail shop is important...

This is APUG small sensors are someone else's problem.

 

[recviemart]

don't forget about size differences, faster means larger.
all you need to know is that you do not need the fastest lens (in general) and low element count lens are better!
gtfoas!
recviem-art.ro

 

[RalphLambrecht]

OK.....I have read a bunch on Wiki, and The Internet in general.
I know the answer is right in front of me, but I cannot find it in All The Text that pops up from a search.

With two of the "same" lens..... For example a Canon 35mm/f1.8 and a 35mm/f3.5.
Lens A = f1.8
Lens B = f3.5
What limits Lens B from being able to open up to f1.8.?

I am kind of wondering...why don't "they" just make all lenses capable of a 2.0 aperture.? Why are/is there frequently the "same" lens manufactured at different speeds.?
Thank You

a couple of things come to mind:
faster lenses need a larger opening with larger glass.
They also need better optical correction to perform as well as slower glad.
all the above make faster lenses more expensive and a lens manufacturer can reach more customers by offering something for every wallet

 

[cuthbert]

don't forget about size differences, faster means larger.
all you need to know is that you do not need the fastest lens (in general) and low element count lens are better!
gtfoas!
recviem-art.ro

Yes the Canon 50mm f3.5 is a very small lens:

This is my 50mm f1.2:

It seems clear to me which one gets more light.

 

[Nodda Duma]

That's an interesting tangent. What was it used for?

Very high speed platforms where heat buildup and impact damage from rain droplets are a concern.

 

[Nodda Duma]

My definition of modern includes the statement lenses Zeiss made pre WWII but did not productionise.
I did not say design I include sales volume per the OP query, which includes cost of Cosina and Leica formed aspheric high volume lenses.

I should have also included the multi coating too as a 4th enabler for the high refractive glass.

Cheap in retail shop is important...

This is APUG small sensors are someone else's problem.

I'm sorry, I'll be sure to put my blinders on and ignore the rest of the design world next time.

 

[CMoore]

Yes the Canon 50mm f3.5 is a very small lens:

This is my 50mm f1.2:

It seems clear to me which one gets more light.

See what I mean.?
...To the Novice/Uninitiated, it makes "you" wonder.....why don't they just have the aperture blades open wider to f2.0.?
If any of you Guys/Gals have a favorite link to a video showing lens construction, I would be very happy to see it.
Thank You

 

[Xmas]

See what I mean.?
...To the Novice/Uninitiated, it makes "you" wonder.....why don't they just have the aperture blades open wider to f2.0.?
If any of you Guys/Gals have a favorite link to a video showing lens construction, I would be very happy to see it.
Thank You

The blue glass in the lower speed lens is limited in speed by the metal holding the glass in position not by the iris.

http://dmc-365.blogspot.co.uk/2012/12/cosina-voigtlander-factory-video.html?m=1

 

[blockend]

As a rule of thumb, it is easier to make a lens of a small maximum aperture produce sharp, well defined images than a lens with a large maximum aperture. Wide apertures translate into exponentially more cost and size for the single benefit of a narrow depth of field.

 

[cuthbert]

See what I mean.?
...To the Novice/Uninitiated, it makes "you" wonder.....why don't they just have the aperture blades open wider to f2.0.?
If any of you Guys/Gals have a favorite link to a video showing lens construction, I would be very happy to see it.
Thank You

Another thing: the FD 50mm f3.5 is a macro lens, so it focuses at short distance and it is also supposed to minimize aberration at short distance, so the elements have to be small (the barrel is big, though) while a fast lens has more glass.

If you compare the FD50mm f1.4 and the 1.2 you'll realise the latter is a scaled up version of the former, but there also is a L version with an extra aspherical element, so generally speaking fast lenses have a more complex formula than a macro or a standard f1.8/2.0 lens.

Of course making a sharp f1.2 is much harder than making a sharp f2.0,that's the reason why fast lenses are considered the pinnacle of optics.

 

[MattKing]

The design of lenses for film formats like 16mm, half frame 35mm, Minox, 110, and disc is an eminently suitable subject for APUG, so feel free to tell us more about the special concerns/opportunities when designing for small "sensors".

 

[Nodda Duma]

Yes, MattKing, yes indeed!

We should also remember vidicon. Entirely analog!

 

[Poisson Du Jour]

I wonder which one is more difficult to make--a good lens or a man-made diamond. And I don't even know if there's such a thing as a man-made diamond.

Crystallised zirconium dioxide (cubic zirconia) is fashioned (by jewellers) into a diamond-like gemstone without the diamond price. It's often described as "a man-made diamond".

Sent from my SM-T805 using Tapatalk

 

[Xmas]

I mean a carbon diamond. I don't know if there's such thing as a man made carbon diamond.

We are off topic and you did not chase links on post #10

But
https://en.m.wikipedia.org/wiki/BARS_apparatus

Yes synthetic girls best friends

 

[Nodda Duma]

Man-made diamond isn't quite off-topic regarding the subject "lens construction" and lens design. Synthetic diamond is used to coat optical windows or lenses exposed to extreme environmental conditions. It's usually called "hard carbon" or "diamond-like coatings" to differentiate from the glittery jewels that make the girls eyes light up.

There are many optical shops who can do these coatings. Reynard Corp. and Jenoptik are two that came up in search. Jenoptik might be a familiar name?

http://www.reynardcorp.com/cleardlc.html
http://www.jenoptik.com/en-diamond-like-carbon-coatings


CVD Diamond works with the material as a substrate and provide optical properties here: http://www.cvd-diamond.com/properties_en.htm

Note the thermal conductivity is way, way better than copper.

II-VI Infrared also makes synthetic diamond windows. http://www.iiviinfrared.com/Optical-Materials/cvd-diamond_substrates.html Obviously the physical properties allow very thin windows, useful for reducing weight of the camera lens / lens assembly.

By the way, CVD stands for "Carbon Vapor Deposition". Basically the diamond window is formed in a vacuum chamber by vaporizing carbon and building up the substrate. Craziness. See https://en.wikipedia.org/wiki/Chemical_vapor_deposition I'm not sure if anyone's made a diamond lens outside a lab. I'd bet that takes a heck of a long time to polish out.




Finally, fabrication of aspheres is often done with .... wait for it ... diamond turning machines https://en.wikipedia.org/wiki/Diamond_turning

Precitech http://www.precitech.com/ produced the first commercial diamond turning machine back in the early 80s (back in the days before digital cameras). They built and delivered SN 01 to the Navy lab that I worked at for several years before moving to NH. The machine was as big as a room. 20 years later, we replaced it with another Precitech machine about the size of a telephone booth. We hoped to donate the original to a University or museum, but no one was interested so it was scrapped.

Diamond turning machines are used to reduce cost of manufacturing of aspheres, making them viable for incorporation in lens design. It's why you started to see the use of aspheric lenses in camera lenses starting in the mid-90s or so. For visible lenses, the microscopic grooves left on the surface by the diamond tool can cause forward scatter aka veiling glare. Aspheric polishing machines have been available for the past few years which can polish out aspheric surfaces on optical glass to avoid that issue... leading to even greater proliferation of aspheric surfaces in camera lenses over the past 5-10 years.


-Jason