Curved Film Plane for Hyperfocal Large Format?

[Michael Heald]

Hello! I have been using a Tachihara 4x5 camera, but I am intersted in experimenting with ULF; I am considering 16x20. I would like to make a hyperfocal comera using a barrel lens such as the long component of a convertible lens. For example, with a 28 inch lens stopped down to f64 would give a hyperfocal distance of about 60 feet, so anything from 30 feet to infinity shoud be in reasonable focus. I plan on making the camera single-shot, that is, loading a single piece of paper or film in the darkroom, then shoot and develop it. Though I would only have one shot, I won't have to worry about ULF film holders and focusing with such a system.

I know that a plain achromat has field curvature much as a pinhole would. How can I tell if an old lens would benefit from creating a curved field plane? Thank you and best regards.

Mike

[Leonard Evens]

I don't know what you mean by "field curvature much as a pinhole would". A pinhole camera doesn't have any field curvature because there is no plane of exact focus. Everything is equally in or out of focus from the lens to infinity. With a perfect lens, geometric optics predicts that the surface of exact focus is a plane. With a real lens, that surface departs from a plane, but how it departs would depend on the specific lens and on the distance at which you focus. If you fix the focus distance at the hyperfocal distance, you wouldn't have to worry about the latter, but you might have to do some considerable work to map out the field of exact focus. Film of course is flat, so while it can be curved to conform to a cylindrical surface, it can't be made conform to another kind of surface such as a spherical surface. A lens not specifically designed for the purpose is not likely to produce a cylindrical field. Also, curving the film will lead to some "distortion" of the image. For a wide angle lens or panaramoic camera, a cylindrical surface's "distortion" may be desirable, but otherwise, you would probably be better off being slightly less well focused with a flat film plane.

[Paul Fitzgerald]

Hi there,

There should be no reason to curve the film plane with a 28" lens at hyperfocal for 16x20 (diagonal approx. 25"). At f/64 or f/90 the depth-of-focus (film side) should cover any difference. The only problems would be focusing the lens at taking aperture, single cell focus shift, and using a deep yellow filter for color shift.

There is a major fudge factor here, I just ran this in a spread sheet based on f/1720, 2" of arc, 2" at 100 yards for the hyperfocal for CRITICAL ENLARGEMENTS. For contact printing the apparent depth of field will be much greater, easier to use. Forget the tables and set the focus to your taste and judgement, at the taking aperture.

I like the idea of single shot to beat the cost of holders BUT this camera will be approx. 20 x 24 x 30 inches large, a bit bulky.

Have fun with it.

[Joakim Ahnfelt]

The reason for curving the film in a pinhole camera is to compensate for light fall off if you have a wide angle camera. To avoid underexposing the edges in other words.

[Walt Calahan]

N.O. Mennescio has it right.

The distance from the pinhole to the center of the image circle is shorter than the distance from the pinhole to the edges of the image circle. Thus, due to the inverse square law, the exposer in the center is different than at the edges. By curving the film plane, you can shorten the distance at the image circle's edge to equal the center.

With extreme wide angle lenses the same event occurs, but we can use centered ND filters to adjust the over all exposer so the exposer is the same across the film plane.

[Michael Heald]

Hello! Thank you for everyone's insight.

I am curious. Any guidance on at what f-stop diffraction would start to be a problem with a hyperfocal camera? Using a cylindrical film plane and slight forward lens tile (if the lens has the coverage) could provide additonal sharp focus so that the lens wouldn't have to be stopped down so much. Best regards.

Mike

[Walt Calahan]

Pardon my typo above

That's "exposure."

What were my fingers thinking as I typed. Must of been in the fixer too long. Grin

[robc]

given your criteria f64 is within tolerance for a coc of 0.1mm which should be fine. That gives depth of focus of + or - 6mm with a 28 inch lens.

www.winlens.de/

download the predesigner software at the above link and have a play. You should be looking at the depth of focus tab (not the systems tab) and can set the coc as small as you want but the smaller it is the sooner you will hit diffraction for your given fstop.

I suggest you check the focal length, object distance and image height to start with and then play play with coc and fstop values watching hyperfocal distance, near and far focus values generated.

[robc]

looking at it again using the software the hyperfocal distance at f64 would be 79.74m with the near sharp point being at 39.87m (for 0.1mm coc). I don't know what is considered an acceptable CoC for 16x20 film but your figures indicate it would about .45mm. Is that what you based your calculations on?

[Michael Heald]

Hello! I was using a CoC of 0.4mm to estimate the hyperfocal distances. Best regards.

Mike