A Site for Photographers by Photographers

Community > Forums > photo.net > Infrared > Formula for finding infared...

Formula for finding infared focus point on a lens?

Carina Cisneros , Jan 06, 2003; 02:06 p.m.

I am wondering if there is such a thing as a formula which would allow someone to calculate the infrared focus point for a lens? Most lenses have this "R" or "mark" on the barrel or in the hyperfocal diagram, but many do not. Is there a way, other than trial-&-error to find this point? If there is such a forumla does it work on all lenses including zoom lenses, or just on primes, or even just on the normal lens for that film format? The question came up first when I realized my Canon GIII does not have such a mark. Also, my M42 lenses of 25mm and 200mm also lack the mark. I was just wondering if there was a rule for finding this. When I look at many lenses with the mark, it doesn't seem obvious to me how its placement was determined, especially on some older zoom lenses. Any info appreciated, even if it is just an answer for the GIII QL17.


    1   |   2     Next    Last

Bob Atkins , Jan 06, 2003; 02:11 p.m.

There is no formula. Every lens is different. It's also likely to be different for long wavelength IR (850nm, Kodak HIE) and short wavelength IR (750nm, Konica 750).

The best thing to do is run some focus tests using YOUR lens and YOUR film with YOUR filter.

Mark Bouquet , Jan 06, 2003; 02:28 p.m.

Actually there is a formula, Bob. On page 53 of "The Camera" by Ansel Adams, "The required adjustment is normally to advance the lens about 1/70 of its focal length..."

Richard Cochran , Jan 06, 2003; 02:38 p.m.

That 1/70 of the focal length is a gross approximation. The actual distance required depends on the dispersion of the types of glass used, among other things. Theoretically, it should be possible to make a lens corrected for chromatic aberration so that a desired wavelength of IR focuses in exactly the same plane as a wavelength of visible light, using the same techniques that are normally used to minimize other chromatic aberrations. I'm not aware of any lens that was designed this way, though.

Bob Atkins , Jan 06, 2003; 02:47 p.m.

That may have worked for simple lenses back in Adams' day but it's not accurate for complex lenses like today's apochromats and zooms.

The IR focus point depends on exactly how the lens is achromatized (if that word exists!). Lenses try to bring all visible wavelengths to the same focus point, but what they do to UV and IR wavelengths is usually unspecified. Many designs will result in significant focus shifts for UV and IR wavelengths. That's not important when what you really want is all visible wavelengths to focus at the same point, so lens designers may not care.

Simple lenses show something like a 1% focus shift (focal length gets longer) due to the change in refractive index of typical glasses between visible light and IR light. Real lenses probably show anything from 0% shift (some lenses are apochomatic all the way out to IR wavelengths) to 2% or even more.

Mark Bouquet , Jan 06, 2003; 02:53 p.m.

Adams does go on to say that some lenses are corrected into the infrared spectrum, and require no adjustment. Out of curiosity, I measured the extension when adjusted to the infrared index mark on each of my 35mm SLR lenses with a micrometer. My 100mm lems extends 1/250 of the focal length. 50mm extends 1/200, 28mm extends 1/155, and 21mm extends 1/150. So I, and Ansel, stand corrected. I guess there is no universal formula.

Bernhard Mayr , Jan 06, 2003; 03:27 p.m.

It's different from lens to lens and depends on how extensively chromatic errors are corrected. Zeiss claims that their superachromat lenses do not need any compensation at all, other lenses do. When looking at the IR index on my CZ lenses, there is quite some variation beween the lenses. There is less adjustment needed with more highly chromatically corrected lenses as judging by number of elements/price/overall MTF performance/difference between tangential and sagittal MTF lines in the corners (D21/2.8, P100/2.0). Simple designs like my tessar 45/2.8 need more adjustment. Focal length has only something to do with it as it is increasingly more difficult to correct those errors with increasing focal length above the normal 50mm lens. Is however also an issue with retrofocus wideangle lenses as Chromatic abberation is also a big problem there. I'm sure Ansel Adams didn't use retrofocus design, that's why he didn't have to take this into account. If it would only depend on focal length wideangles should not have any problem, which is far from true. Lens makers have to go a long way to correct chromatic abberation in retrofocus wideangle lenses.

David Vatovec , Jan 07, 2003; 02:07 a.m.

Canon says that their "L" lenses don`t need any correction for IR focus (that is true for the 135 f2.0 which i have - not sure about all the other "L" lenses).

Winfried Buechsenschuetz , Jan 07, 2003; 04:35 a.m.

It's not completely a matter of correction of chromatic errors. If a lens is achromatic or apochromatic this refers to the VISIBLE range of the spectrum only. On the other hand the near infrared band is close to the low (red) end of the spectrum so you should not expect too much focus difference.

Also, there are two types of chromatic aberration. Even if the lens is sufficiently corrected for longitudinal chromatic aberration (i.e. focus difference) there might still be a significant amount of chromatic magnification error, i.e. objects in the infrared range may have the same focus BUT the size of the image is slightly different. This might cause blur on infrared pictures.

To make a long story short I agree with previous posters that it is hardly predictionable how a lens will behave in the infrared range.

Milos Bozovic , Jan 07, 2003; 04:39 a.m.

Try using smaller apertures around f5.6 or f8 for landscapes or cityscapes. That will do in most cases. On the digitals even apertures of f3.5 or f4 should be sufficient. The question is only how the CCD of the G3 should capture the infrared ligth?

    1   |   2     Next    Last

Back to top

Notify me of Responses