Close up DOF with LF

Given ARCA-SWISS 4X5 Field Camera and 210mm lens, is there a way to

expose an image where front of lens to subject is 30 inches and

desired depth of field is 12 to 24 inches? For example, small field

of slate (rock) sheets lit by normal sunlight from 1 o'clock

direction? Experience has shown only about 2 inch DOF with front

tilt not adding much. Back up and crop? Get new lens? Find

different subject? Thoughts? Tricks?

I assume you mean that you want a depth of field of 6 to 12 inches on either side of the plane of focus at 30 inches. It will be hard enough with 6 inches, so I will concentrate on that. If I haven't understood you, please explain, and I will do the calculations over again.

 

Let me go through the reasoning without assuming a tilt, so you can check the arithmetic to make sure I got it right. If you want to skip the calculations, jump ahead to the answer.

 

The magnification is related to the subject distance by the formula

 

sub-dist/foc-len = 1/mag + 1.

 

30 in = 762 mm, so the quantity on the left is 762/210 ~ 3.69, and the mag is the reciprocal of that less one or 1/2.69 ~ 0.38.

 

For closeups, the formula for depth of field on either side of the plane of focus is

 

f-number x coc x (1 + mag)/(mag squared)

 

Taking coc = 0.1 and mag = 0.38 reduces this to

 

f-number x 0.96

 

You want this quantity to be 6 inches or 152 mm. So the f-number would have to be

152/0.96 or about 159.

 

Stopping down to f/159 is not feasible with any lens I have, and if it were, diffraction would kill the image. Note also that some people wouldn't find a coc of 0.1 mm adequate and that would increase the needed f-number even more.

 

Of course, if the subject is very close to planar, you may be able to get everything in focus by using a tilt. But if there is much vertical variation in the subject, that won't help. I can do the calculations for that too, but it is more complicated, so perhaps you should describe what you want more precisely.

 

You would do better with a shorter focal length lens. A 135 mm lens should do it for 6 inches of DOF on either side of the plane of exact focus, without tilting. Stopping down to f/64 should work.

<i>You want this quantity to be 6 inches or 152 mm. So the f-number would have to be 152/0.96 or about 159.<p>

 

Stopping down to f/159 is not feasible with any lens I have, and if it were, diffraction would kill the image. Note also that some people wouldn't find a coc of 0.1 mm adequate and that would increase the needed f-number even more.</i><p>

 

This is absolutely feasible, but not with a lens. For this kind of depth of field, you need to use a pinhole. A simple shutter will suffice, exposures will run many seconds, even minutes, at effective focal ratios between f/250 and f/450 in an equivalent to the 210 mm focal length you specify. Everything in FOV, from an inch in front of the pinhole to the horizon, will be just as sharp as everything else, and on a 4x5 for contact printing, probably sharp enough not to elicit comment (though even slight enlargement will very quickly reveal the "everything is slightly, but equally soft" character of all pinhole photography).<p>

 

Best part of all, you can make a pinhole youself from a used soda can, unscrew the lens elements from one of your existing shutters, and spend nothing to get this incredible depth of field!

It isn't clear from your description how the the rocks are arranged. If you have flat rocks (suggested by the slate) laying on flattish ground, then changing the plane of best focus with a front tilt (or back tilt) might greatly reduce the amount of dof required to have all of the subject in focus. If you can find a plane that minimizes the greatest distance of the subject from that plane, you will reduce the dof needed to get all of the subject in focus. This will work if your lens has enough coverage to permit the amount of front tilt required to focus on that plane. Alternatively, you can use back tilt, but this will also change the perspective.

 

There must be previous threads on this subject. A complication is that with the front tilted, the dof will be approximately wedge shapped, being larger about more distant portions of the plane of best focus.

 

If camera movements don't make your intended photo possible, then you can try a new composition, or back up and crop, or try different subject, etc., as you suggest.

 

Lighting doesn't have much to do with dof issues, unless some areas are off the brightness scale (e.g., in deep shadow) so that it doesn't matter whether they are in focus or not. Nor does the particular camera body, as long as it allows the movements needed to reach the desired plane of focus.

Donald,

 

You are of course correct about the use of a pinhole. But, as I noted, at least in some circumstances, some of us would find the diffraction induced fuzziness unacceptable.

Really interesting answers. Thanks, folks. To clarify a bit, the slate sheets are verticle and sheered off just above ground level. The ends of the sheets form irregular lines across the 12 to 18 inch square of ground that I'd like to photograph with the camera aimed down at about 35 degrees and out about 30 inches from the nearest slate. Using the mm scale on my extended rail, I get about 30mm difference between the near and far "in focus" locations, which when multiplied by 5 gives me a minimum F/stop of about F/150; not far from what the F/159 calculated by (Leonard? - Can't see the names from this window). Anyway - thanks for the thought processes so far. If there are any more, please feel free to pass them on. They're very helpful. (Note to file: save one diet coke can this week.) :-)

I think I get the picture now---pun intentional. I assume you want to point the camera down and focus on the middle of the region you want in focus. I would guess then that the lens would be about 900 mm from the middle of the region. That would place the hinge line about 665 mm away from the lens when it is in neutral position. A 210 mm lens at f/64 (using a coc = .mm) has a hyperfocal distance of about 6890 mm. At that distance, the range in focus about the plane of exact focus would be about 665 mm in either direction. If I estimate the near point at about 750 mm, that yields about 72 mm in focus on either side of the exact plane of focus at that distance. Using 600 mm, the number is about 58 mm. So whether or not you can get everything in focus with a tilt depends on the vertical extent. If it is less than about 4 inches altogether, you ought to be okay.

 

But my estimates may be off. The rule I'm using is that the variation on either side of the plane of exact focus at the hyperfocal distance is approximately equal to the distance from the lens to the hinger line. At other distances you modify proportionately.

Perhaps the photo would look good if you used front tilt (and possibly swing if you aren't looking straight at the slates) to place the plane of best focus approximately through the tops of the slates and stopped down enough to get all of the tops in focus, but accepted the ground itself not being in focus, rather than focusing midway between the ground and the tops of the slates and trying to get it all into good focus. People are used to seeing distant parts of a photo out of focus, and if the tops of the slate are more interesting than the ground, my inclination would be to emphasize the tops of the slates in my focusing and allow the background to be not in perfect focus. Having the background out of focus might help the photo by making the ends of the slates standout more -- you might experiment with several apertures. Whether this idea will work depends on the objects and what YOU want.

Leonard and Michael: Thanks for the additional input. I'm going to try and upload a photo taken with my cell phone of the LF set up I was taking. You maybe already have a good enough mental picture, but I just want to try and do the process. Thanks again.<div></div>

Oops, sorry. Wanted to also tell Leonard that I downloaded his paper on the algebra involved in lens focusing. Learned "hinge", "coc" and a few other terms I'd never seen. Still lots more to read. That paper is a massive headache to read for someone that's been out of math class for about forty years. :-)

Many process camera lenses have diaphrams that can be stopped down alot. Our 600mm apo ronar has stops below the usual working F22 of F32; F45; F64; F90; F128; F180; F260. These fstops are actually on the lens barrel.

Unfortunately there aren't any good sources of information about these matters which are both clear and reasonably quantitative. You might look at some of the links at www.largeformatphotography.info. Several discuss the details of tilted subject plane, but I think only three discuss the depth of field about that plane. Those are Bob Wheeler's, Merklinger's, and mine. Merklinger's descriptions are probably the least equation ridden, but I never found them very clear, and they are spread out in several different documents.

Leonard: I'm working through your document. That should be sufficient. I'm also using Stroebel's "View Camera Technique" (7th Ed.) to supplement and further explain in terms of definitions and drawings. With both of them I'll be able to learn enough and figure it out. Thanks for all the help.

Some DOF calculations are approximations; and fall apart more at close distances. For macro work; be carefull and do some real world tests too.

Kelly, Thanks for your comments. I actually was looking up information about "process" lenses after your earlier post. More new knowledge. Thanks again.

Donald: Just to flail the pinhole horse one more time (which I'm apt to do), my experience with "optimized" pinhole cameras using LF sheet film (ie, f/ratios that conform to the Rayleigh limit) in closeup applications is that you tend to lose image sharpness at extreme closeups. At 30 inches you may not see this effect, but much closer and you will.

 

I don't attribute this loss of sharpness to the fact that the subject is running out of micro detail; most of my test subjects have enough micro detail that should show up at extreme closeups.

 

Rather, the geometry of ray tracing shows that linear perspective induced edge fuzziness increases faster than diffraction effects at extreme closeups.

 

For this reason, it is advantageous to use a pinhole f/ratio smaller than Rayleigh's recommendation (which is optimized for objects at infinity) for applications primarily requiring extreme closeups.

 

A pinhole image of a landscape/still life at/near infinity can be amazingly sharp, considering the smallest object that can be imaged is a bit larger in diameter than the pinhole itself. Think of detail on distant tree trunks and rocks around .2mm in size and up; there's a lot of texture and detail in that size range to give a sense of sharpness. Its not as good as an image from good glass, but such an image, when contact printed, has a grainless smoothness that can be quite appealing.