PDA

View Full Version : Measuring FL - at rear lens or in between the lenses?

renes
17-Jun-2010, 10:59
I have a brass Aplanat Extra Rapid L.Renaux Suter f/2-100 with engraved 17cm focal lenght. Looking on the GG I felt it was a bit longer focal than marked. So I measured the FL from the GG and 170mm ended at the rear lens instead in the middle of the front and rear cell (and here is about 190mm).

My petzval 150mm reaches 150mm exactly in between the cells.

How do you explain it?

Paul Metcalf
17-Jun-2010, 11:10
I have a brass Aplanat Extra Rapid L.Renaux Suter f/2-100 with engraved 17cm focal lenght. Looking on the GG I felt it was a bit longer focal than marked. So I measured the FL from the GG and 170mm ended at the rear lens instead in the middle of the front and rear cell (and here is about 190mm).

My petzval 150mm reaches 150mm exactly in between the cells.

How do you explain it?optics (gaussian), refraction, wave theory of light, magic :p

renes
17-Jun-2010, 11:19
optics (gaussian), refraction, wave theory of light, magic :p

or wuwuzele :D

Paul Metcalf
17-Jun-2010, 12:15
Well, it was a serious answer (maybe not the magic part), but you have to do the hard work of reading up on the three other topics (gaussian optics, refraction and wave theory of light). Gaussian optics will describe what the principal points/planes are of a "thin" lens, refraction will tell you how to construct a ray trace of your more complex lens "system" (the aplanat or the petzval made up of one or more "thin" lenses) using these principal points/planes, and the wave theory of light will tell you what the index of refraction is for the medium (glass/air) the light is passing through (which is a function of wavelength). A good understanding of calculus helps. (I'm assuming you're shooting wide open, otherwise you'll have to also read up on diffraction - through the aperture).

Or, back to magic, you accept that these things are designed intentionally for an intentional purpose, and hopefully the designer has told you what the focal length is, and the rest is up to you to make "magic" of it. If you aren't told what the focal length is, then what you have done to estimate it is a good approach, in that you have a reference point from which to measure magnification (or bellows factor) if you're doing close up/macro shots.

17-Jun-2010, 13:39
Focal length is the distance from the image plane to the "rear nodal point" of the lens when the lens is focused on an object at an infinite distance. Depending on the design of the lens, the rear nodal point might be ahead of (in the case of a "telephoto" design), behind (for a retro-focus or inverse telephoto design), or near the center of the lens (for conventional designs.)

Telephoto designs are often used to make the lens more compact or to be compatible with limited bellows extension. Retro-focus designs are often used with SLR cameras where the lens must stay clear of the moving mirror. Most lenses used with view cameras (plasmats, tessars, dialytes, dagors, double gauss and even biogon-like super-wides) fall within the conventional category, but even then the location of the rear nodal point is likely to be somewhat ahead of, or behind, the center of the lens. Depending on the design, estimating the focal length by measuring from the film plane to the lens is either unreliable or completely misleading.

The topic of measuring focal length comes up frequently. Here is a good discourse from one of the best informed commentators on the subject.

http://www.largeformatphotography.info/forum/showpost.php?p=584863&postcount=2

renes
17-Jun-2010, 14:11
Thanks. That's a new field of theory for me.

corgan4321
24-Jun-2010, 19:16
So if focal length is the distance from the image plane to the "rear nodal point" of the lens when the lens is focused on an object at an infinite distance, with every single lens, focusing on a subject CLOSER than infinity will always require a greater amount of bellows extension?

So, what attribute of a lens dictates the curve the lens will follow in terms of distance from subject to necessary bellows extension?

Now, what attribute of a lens specifies how close the lens will be able to focus? Simply how close the nodal point can get to the film plane? So maybe the lenses coverage is what dictates this. For example, my Rollei TLR can only focus up to about 2 feet away. Could this be because if the nodal point of the lens got any closer to the film plane, it wouldnt cover the full 6x6cm?

rdenney
25-Jun-2010, 05:45
So if focal length is the distance from the image plane to the "rear nodal point" of the lens when the lens is focused on an object at an infinite distance, with every single lens, focusing on a subject CLOSER than infinity will always require a greater amount of bellows extension?

So, what attribute of a lens dictates the curve the lens will follow in terms of distance from subject to necessary bellows extension?

Now, what attribute of a lens specifies how close the lens will be able to focus? Simply how close the nodal point can get to the film plane? So maybe the lenses coverage is what dictates this. For example, my Rollei TLR can only focus up to about 2 feet away. Could this be because if the nodal point of the lens got any closer to the film plane, it wouldnt cover the full 6x6cm?

Yes, focusing closer requires more bellows extension. The nominal (as in, named) focal length is at infinity, but the actual focal length varies by the distance to the subject.

There is no practical limit to how closely a lens can focus. All you need is the ability to rack it out far enough. In theory, putting the lens far enough away from the film would get you into the microscopic world, but there are a couple of limitations. One is that the design of the lens may interfere with the subject--it is possible to focus on the dust on the front surface of the glass with some lenses. Another is that few lenses designed for use on cameras are sufficiently optimized for situations where the projected image is thirty times the subject size, or more, and the image quality would completely break down (reversing the lens can, in some measure, correct this). And then there's the issue of speed--the image would be so dim, you would have a hard time focusing it, like too little butter spread over too much bread.

The way the math works, a 150mm lens will be 150mm from the rear nodal point to the film when focused at infinity, but 300mm from the film to the rear nodal point when the image on the film is the same size as the subject in real life (i.e., 1:1).

The reason Rolleiflexes focus no more closely than they do is because of their mechanical design. Also, the double-gauss lenses used on Rolleiflexes after the early 50's are optimized for more distance subjects and don't perform as well when subject magnifications reach 1:1. I used a Mamiya C-series TLR for many years in what little commercial work I did. It has rack-and-pinion focusing with a connecting bellows rather than the cam focus of a Rolleiflex, and could focus down to 1:1 with some lenses. But good luck trying to understand what the taking lens was seeing. At 1:1, the viewing lens of a TLR seems like it is about a mile to one side. Many rangefinder cameras also limit the closeness of focus because of the mechanical limitations of the rangefinder. One of the key advantages to the single-len reflex design (and to viewing on ground glass at the film plane) is that you can see exactly what the lens sees. That's why all SLR manufacturers offered purpose-built macro lenses, supplemental bellows for close focusing, and even reversing rings for their normal lenses.

Rick "mechanical design limits how closely it will focus, optical design limits how sharp it will be when focused" Denney

Helen Bach
25-Jun-2010, 05:48
So if focal length is the distance from the image plane to the "rear nodal point" of the lens when the lens is focused on an object at an infinite distance, with every single lens, focusing on a subject CLOSER than infinity will always require a greater amount of bellows extension?

Yes.

So, what attribute of a lens dictates the curve the lens will follow in terms of distance from subject to necessary bellows extension?

The focal length. There are a few ways of writing this, but the simple, easy to remember version is

1/FL = 1/do + 1/di

FL = focal length of lens
do = distance from object to front, or first, nodal point of lens (a lens has two nodal points - they may be coincident)
di = distance from image to rear, or second, nodal point.

Now, what attribute of a lens specifies how close the lens will be able to focus? Simply how close the nodal point can get to the film plane? So maybe the lenses coverage is what dictates this. For example, my Rollei TLR can only focus up to about 2 feet away. Could this be because if the nodal point of the lens got any closer to the film plane, it wouldnt cover the full 6x6cm?

Remember that the closer you focus, the further from the film plane the rear nodal point is (hence the image circle gets larger). The closest you can focus is, therefore, determined by the maximum distance that can be achieved between the rear nodal point and the film plane. It's limited by the bellows draw or the range of the focusing helicoid etc.

The limit of close focusing in terms of the lens only is the focal length of the lens - the lens cannot focus on an object that is closer to the front nodal point than the focal length of the lens. If it was closer, the image would be formed beyond infinity (ie it would be a virtual image in front of the nodal point, not a real image behind it).

Adding supplementary lenses produces a combined lens with a shorter focal length than the main lens, enabling the combined lens to focus on closer objects at the same bellows extension.

Best,
Helen

25-Jun-2010, 06:56
I beg to differ with Rick in that the Focal Length of a (non-zoom, non-varifocal) lens is a constant and does not change with the distance to the subject. The Image Distance changes in accordance with the equation quoted by Helen (and the change in Image Distance will affect the Field of View), but the Focal Length is a constant. If the Focal Length changed with subject distance then Helen's equation would not be usable. But it is useful, and in fact is fundamental.

Rick seems to be using the term "Nominal Focal Length" to refer to this constant. The term "Nominal Focal Length" is usually used to distinguish between to Focal Length that a lens is marketed under, say a 210mm Schneider Symmar, and the actual or Effective Focal Length of the lens as designed and manufactured, which is often a bit different. For example, Schneider's Vintage Lens Data (https://www.schneideroptics.com/info/vintage_lens_data/large_format_lenses/symmar/) shows that the "Effective Focal Length" of their old convertible 210mm Symmar is actually 212mm.

For purposes of choosing which lens will give you the subject distance or field of view you want in a given situation, the Nominal Focal Length is adequate, but if you were grinding a rangefinder cam or building a fixed focus box camera, then you would probably want the actual or Effective Focal Length.

Hope this sheds more light than heat - Alan

Emmanuel BIGLER
25-Jun-2010, 10:12
Hello all !
Regarding different methods to find the focal length of an unknown photographic lens, I had mentioned 3 methods (thanks to Aducanson for digging successfully inside the archives)
http://www.largeformatphotography.info/forum/showpost.php?p=584863&postcount=2
which can be considered as "absolute" methods.
But there is another one that we discussed, a comparison method, giving a very precise value of the unknown focal length. You need to compare with another lens, for example the "212mm" Schneider as quoted above.

You measure on the ground glass the size of the image for a far distant object for both lenses. A simple ruler is enough. The ratio of the focal lengths is equal to the ratio of the image sizes.
Hence, if the unknown lens is a 170mm, compared to the 212, the image size given by the 170 will be slightly smaller than for the 212, the proportion being equal to 170/212.
WIth a simple ruler you can measure a distance with a precision better than 1% (this is a precision of one mm over 100mm, easy !), hence if the reference focal length is known (e.g. by looking into the manufacturers' archives), you get an estimate of the unknown focal lengths within a few percent.
Not precise enough, may be, for photogrammetry, but a few % on the value of a focal length is true luxury for fine art photography ;)