Aperture bigger than F1 ?

[gandolfi]

well - I recently heard two opposite points on the question on the F-scale.
And if it is possible to reach a bigger aperture than F1 or not..

(I know there's lenses with aperture numbers less than 1, but I'll get back to that.)

First: you can easily make apertures larger than 1 - just make the lens big enough...

second: no - if an aperture has F:1 as it's biggest, there's no light loss from outside the camera to the film...

the reason you can make a F:0.95 or "lower", is that you can use special glass that allows light rays, not visible to the eye to hit the film.

I found that I got confused when I heard this.
and I have tried to find out the "truth", but to no avail...

I know what I believe, but if I am wrong, I have been teaching the wrong lesson for many years...


Anybody know? And can anybody explain, so a simple minded guy as me, can understand?

 

[edp]

Stanley Kubrick's f/0.7 lenses:

http://www.visual-memory.co.uk/sk/ac/len/page1.htm

 

[Lars Jansen]

Well, it's easy to get an F-stop smaller than 1. As you may or may not know, you calculate F-stop by dividing the focal length of the lens by the diameter of the opening. This means that a 50mm/1.0 has a maximum opening of 50 mm. This also means that a 50mm lens with a maximum opening of 100mm would be a 50/0.5. So the first statement is true.

Lars

PS. Check here for more in-depth info: http://en.wikipedia.org/wiki/F-number

 

[Removed Account2]

f-number is dead easy focal length divided by lens diametre.

50mm lens, 2 cm diametre : 50/20 = 2,5 i.e. f:2,5

50mm lens 5 cm diametre : 50/50 = 1,0 i.e. f:1,0

50mm lens 8 cm diametre : 50/80 = 0,625 i.e f:0,625

There are few lenses with larger aperture and smaller f-number than 1,0
Leica has had a few I think, Canon had their famous f.0,95 on their Canon 7s.

What confuses you might be the THROAT. What the light has to pass after the lens and before hitting the film.

Most times the throat is not wide enough to allow the necessary big lens elements deep inside to allow the light to bend enough to reach the corners.

Thats why the Canon 7s had an outside BAYONET as an extra to the Leica 39mm screw thread, allowing the engineers to design a larger back element.

As you can see there is no magical border here, just practical (and impractical) limits.

When Canon went to their bayonet SLR lenses, which ended with their FD lens series, they went for a bigger throat.
When Canon went to their EOS bayonet, they went for an even bigger throat.

In hindsigt the designers have always been too short-sighted.
We want more, always more.

 

[fdisilvestro]

The theoretical maximum aperture limit is f 0.5

 

[Removed Account2]

Well not so sure about that, they use some incredible lenses inside computer chips factories I'm told, I've read somthing ages ago, but forgotten...

But of course if one draws out in scale how a f:0.5 lens looks as seen from the side, there is at least a practical limit.............

 

[holmburgers]

I'm not sure about any invisible rays playing a part. Those rays (UV and IR presumably) would be there whether or not the aperture is f/1 or wider and no special glass is used in making wide aperture lenses, as far as I know. That sounds like a red herring.

 

[Q.G.]

Larger than f/0.5 and the sine-condition cannot be met, and there is no "image quality" to speak of.

 

[fdisilvestro]

I'm not really a optical expert, but I have read something about. From the book "Applied Photographic Optics: lenses and optical systems for photography" by Sidney F. Ray :

"It has been shown that the maximum theoretical aperture of an aplanatic distortion-free lens on-axis, assuming no transmission losses, is f/0.5"

Page 288, you can check some pages of this in Google book here

 

[Removed Account2]

Well but note the qualifiers : Aplanat, distortion free, loose that and where is the limit? That is not a theoretical limit, its a practical limit.

 

[Q.G.]

That is not a theoretical limit, its a practical limit.

If you think something intended to be an image forming system not being able to be an image forming system due to some constraints a practical matter, it is.
If you would want to consider why it cannot be an image forming system however...

 

[ic-racer]

It is a ratio of two lengths, there is no limit to the number. Practical concerns for glass lens construction were pointed out above.

 

[Q.G.]

It is a ratio of two lengths, there is no limit to the number. Practical concerns for glass lens construction were pointed out above.

That's not right.
There is a limit, and it has been pointed out what that limit is.

 

[gandolfi]

thamks all - I am getting wiser all the time.....

 

[Jeff Kubach]

I woundn't mind having a fast lens but those are too fast for my tast.

Jeff

 

[Bob-D659]

One limit of aperture size is the difference in the refractive indexes of air and glass. Some microscope lenses get around this problem by coupling the lens to the subject by immersion of both in various liquids having a refractive index up to 1.5 or higher.

Bob

Edit: should have added this link for you. http://en.wikipedia.org/wiki/Numerical_aperture

 

[keithwms]

The f/0.5 limit must be for a very specific design and assuming no losses.

But anyway, you can see that the fastest lenses on the market right now are close to f/1. I sometimes use a 1.2 and wouldn't really know what I'd do with anything faster!

Well not so sure about that, they use some incredible lenses inside computer chips factories I'm told, I've read somthing ages ago, but forgotten...

You might be referring to superlenses, a topic I worked on, in my former life. To make a long story short, the expectation is that you can dose photoresist with resolution many times finer than the wavelengths of the usual deep-UV lithographic light sources e.g. getting 20 nm definition from a 157 or 193 nm wavelength, which would be a game-changer for full-wafer processing. Making this possible is a pretty big push right now in the electronics industry. A separate but related thrust is UV immersion lithography, which is already in use. In addition, near-field 'lenses' have been with us for quite a long time but suffer from substantial throughput issues.

 

[Removed Account2]

Keith that link gave me a headache! Diff math that goes on a full line after a full line usually does that.
I'm more like a Reagan guy in this respect, give me a one page synopsis of a 2000 page, substantiated report!

 

[ic-racer]

BTW two F 0.9 Bolex lenses on ebay right now. Only 2/3 stop slower than Stanley...

 

[Photo Engineer]

the reason you can make a F:0.95 or "lower", is that you can use special glass that allows light rays, not visible to the eye to hit the film.

I found that I got confused when I heard this.
and I have tried to find out the "truth", but to no avail..

This part is really rather bogus! If you cannot see the light it is either UV or IR. If it is either of those, well, all non-IR films are IR blind, and all films have UV overcoats (as do most lenses) to prevent distortion of the image from the inherent UV sensitivity of all films.

PE

 

[gandolfi]

This part is really rather bogus! If you cannot see the light it is either UV or IR. If it is either of those, well, all non-IR films are IR blind, and all films have UV overcoats (as do most lenses) to prevent distortion of the image from the inherent UV sensitivity of all films.

PE

right - that's why I got confused, and wanted some good info.

 

[Jim Jones]

Of historical interest only is the Leitz Summar f=7.5cm 1:0.85 of 1934, near the bottom of http://www.taunusreiter.de/Cameras/Biotar_en.html (text in German). The photo and a text in English appears on page 61 of The Leica and the Leicaflex by Theo Sheerer and Josef Makovec.