PDA

View Full Version : Lens Coverage Test for 16x20 Hyperfocal Camera



Michael Heald
18-Dec-2005, 15:54
Hello! Previously, I asked for help in constructing a hyperfocal 16x 20 camera. Assuming a CoC of 0.4mm, and f64, using the 28 inch front cell of an old Gundlach Rapid Rectilinear Convertible lens, the hyperfocal distance was 70 feet, with the near focusing distance of 35 feet. I plan on loading the camera with film/paper in the darkroom for single shot use, and use it for landscapes.

Today, I used a large cardboad box, and taped 8x10 VC paper to each corner, trying for the appropriate film plane of about 29 inches from the lens. I curved the film in the right upper quadrant and left it flat in the left lower quadrant. I exposed for 40 seconds, which appears to be overexposed about 1 to 2 f-stops after development.

Happily, I see no light falloff at the corners. The curved film plane didn't seem to do much with light fall off, since the edge was about 85% illuminated compared to the center at this focal distance, but I'll probably keep it in the final design, since it would help with the focus at the edges of the film (I think).

Anyway, the 28 inch cell of the Gundlach convertible seems to do the trick. I'll probably purchase an inexpensive shutter and attach it to the front of the lens cell, since at f64, exposure times may be too short to control accurately with a lens cap.

Best regards.

Mike

Dan Fromm
18-Dec-2005, 17:35
Pardon my ignorance and naivete, but why did you choose a CoC of 0.4 mm. If you're going to contact print, wouldn't .1 mm be safer? I mean, if you're getting only 2.5 lp/mm on distant objects, won't even a contact print be visibly soft unless you can force a long enough viewing distance?

Cheers,

Dan

Craig Wactor
18-Dec-2005, 19:24
cool project. I hope you are planning on making one out of something more substantial than cardboard. I have been "planning" on making a box camera like this for a while now.

Jay DeFehr
18-Dec-2005, 22:24
Dan,

I don't mean to speak for Michael, but the optimum viewing distance for a 16x20 print is 4X as distant as that for an 8x10, so if .1 is good for 8x10, .4 should be good for 16x20. As for "forcing" viewing distance, the same could be said for the use of loupes.

Jay

Michael Heald
19-Dec-2005, 04:23
Hello! I used 0.4mm for the CoC based on the recommendations from "View Camera Technique" which is based on the anticipated viewing distance. I don't have enough experieince to tell if a smaller CoC would have been better.

I'll probably build the camera with 1 to 2 inch thick styrofoam sheets. This comes in a variety of weights/strengths, though I'll use the low weight to start with, that way, even though the dimensions of the camera/box will be large, the weight will make it relatively manageable. Best regards.

Mike

Michael Mutmansky
19-Dec-2005, 07:22
Michael,

I recommend you make the camera out of black foamcore sheets, not styrofoam. It will be light, inexpensive, and will be very easily fabricated. You can double up on the places where you are concerned about stiffness, and you can easily make a curved film plane by shimming a sheet into position with curved wedges.

Get some hot glue for fabrication, and then tape all the corners with black gaffers tape. It'll be absolutely light tight (as long as you design it correctly) and fairly light and probably not as bulky as styrofoam. You can get a case of black foamcore from various framing suppliers for around $65.

I don't think the logic for the COC holds true for two reasons. First, the viewing distance is open somewhat to intrepretation. The viewing distance will not necessarily be 4x the distance of an 8x10, because at 2x the distance, a 16x20 will occupy exactly the same FOV as an 8x10.

Second, the notion that the COC calculations are for typical viewing distances is somewhat idealized. I think they sould be considered MINIMUM viewing distances, not typical viewing distances. Even though a typical viewing distance may be 2X the 8x10 distance, the viewer can still approach the image in the same manner as a smaller print unless you put a piece of furniture in front of it. So, if someone is 'drawn in to' the print, the image will fall apart as they approach.

If you are comfortable with this, then continue on, but if you want a print that will hold up under some inspection, then you may need to think about dropping back to .1mm.

Good luck on the project. It's a fun thing to do, and can prove to be a useful and mind opening project on multiple levels.

---Michael

Dan Fromm
19-Dec-2005, 07:32
Another innocent, naive question.

Why curve the film plane? Y'r lens is intended to image a plane on a plane, not something curved on something else curved. A curved film plane seems a recipe for a reasonably sharp vertical strip in the center of the image and fuzz elsewhere. If that's what you want, fine.

Michael Heald
19-Dec-2005, 07:54
Hello! Thank you for the info about the foam core. I'll look into it.

As far as the CoC, using 0.1mm pushes the hyperfocal distance to about 280 feet, so anything within 120 feet would be out of focus, at least at f64. I was concerned that pushing to f128 would start to give unacceptable diffraction results. It seems like a trade off either way. I can experiment a bit with the film plane distance once the camera is constructed, in order to see which works.

Good point about the curved film plane. Since this is an old lens, I thought it would suffer from light fall off and curvature of field. Curving the film plane would minimize these two possible effects, at least in the horizontal film dimension. Is ther a way to check for these aberrations in my old lens?

I'd appreciate some other thoughts as to the pros/cons of using a curved film plane in this situation. Best regards.

Mike

Michael Mutmansky
19-Dec-2005, 08:34
Michael,

In particular with a converted lens, you probably will have field curvature, but depending on how much of the FOV you are using, it may not be very significant at all, and may be better (simpler) to use a flat plane for the film.

Because you are calculating for hyperfocal distance and using a relatively small aperture, I don't think a curved film plane will be necessary unless the front element has a particularly strong field curvature.

You can test that a bit by shooting at wider apertures, focused on a specific distance (a building wall). If you are able to get reasonably sharp images at the specific focused point distance at a larger aperture, then curving the film plane will not improve the performance substantially.

If you use a flat plane, you can adjust the focus point to split the difference between the corners and the center point. That will make a much larger area remain at the idelized focus distance, which will improve corner sharpness, with little to no penalty in terms of center sharpness.

With the lens you are using (slightly longer than normal), I doubt you need to worry about field curvature at all. Idealized, when focused at 29" in the center, the corners are at 31.7 inches from the lens. if you bumped the focus point in to about 28.25" or 28", your average distance for the entire sheet of film will be close to the idealized 29".

---Michael

Dan Fromm
19-Dec-2005, 09:01
If you are contact printing, f/128 should be fine. At that aperture the lens should give you njo less than 10 lp/mm on film, and that's about what you need. If you plan to enlarge, reconsider what you're trying to accomplish.