Sharpness and wavelength

[Rich Ullsmith]

I have been making phole images for a few years, and have several home-made boxes. I projected all the pinholes on the wall, got an exact measurement of size, did the math to figure out the optimum "focal length," and inspected the pinholes under a microscope for any defects.

I'm happy with the images, yet they don't appear as sharp as some others I see.

Talking physics with the neighbor, he said that I would get a sharper image if I filtered out one of the primary colors. He said since I have all these different wavelengths squeezing through the pinhole and overlapping, if some were eliminated the result would be a sharper image.

Well, I do that already. I shoot mostly landscape and cloud movement, so I have R25 filters on everything.

But. . .since I'm filtering the shorter wavelength part of the spectrum (blue), and allowing the longer wavelengths (red) to pass, is this the cause of a less-than-sharp image? I guess that is my question: when using a pinhole aperture, will shorter wavelengths produce sharper images than longer ones?
If so, I need to start culling 58mm polarizers from the junk sales.

Love to hear some ideas on this.

 

[Ole]

The pinhole should be optimised for the wavelength you want to use; there is no physical law that says that one is inherently sharper than the other.

Well - when you get right down to the microscopic scale, shorter wavelengths will produce a sharper image than longer ones - or at least allow greater magnification. That's why they make electron microscopes. But that's totally irrelevant to pinholes.

 

[Rich Ullsmith]

Perhaps by filtering the shorter wavelengths, I need to recalibrate the optimum distance of pinhole to film plane? This would be much easier for me, since the film holders are adjustable. Remanufacturing and measuring a new set of pinholes would not be fun for me.

I think the formulas I used assumed a wavelength in the middle of the visible spectrum. Maybe I need to recalculate, shift that number to the longer end.

 

[Greg_E]

Pinhole Designer has a box to change the wavelength of the light, it default to a middle of visible spectrum. I tried a few different filters, and didn't see much difference when I was playing around. Though if you want the sharpest, you might want to calculate for blue light, and use a strong blue filter. Also put a UV blocking filter on it. Many people suggest a user constant in Pinhole Designer of 1.5 to 1.56

And sharpness can sometimes come down to the size of the film you use. If you use 35mm, it must be enlarged by a lot to have an 8x10 print, but if you use 8x10 film, there is no enlargement. Looking at images on the web you are normally looking at a size reduction too, especially if it is coming from a large format like 8x10, this will increase apparent sharpness.

 

[Rich Ullsmith]

You know, I didn't think of a UV filter but that would make sense with unfocused light. And I have more than a few laying around.

Can't use a blue filter, because it goes against my sky/cloud separation. I have a couple 58mm polarizers, I'll compare that to the R25 next time I get out. BTW, it's 4X5 sheets, and I'm not saying it's not sharp. Just not like some I have seen.

Someday I will learn how to scan and post.

 

[Greg_E]

It seems to be a long journey towards "perfection". I've only just started, but the information above came mostly from Jim Jones and a few other people at www.f295.org

In my limited testing I did try a few different things including putting filters over the light source, and filters over the "lens". And they made little difference. With the dark red filter over the "lens" you should be getting minimal UV to the film, and a lot of film already rejects UV light, but I still wonder if it impacts things in a negative way. However the IR light might be messing with the sharpness, so you might try an IR blocking filter.

I never did try a polariser over the pinhole, that might be real interesting. I'll have to try that and see what happens. It's possible that polarized light could increase sharpness a little, and would be worth a try.

 

[Tom Miller]

Try making a photo without any filtering; the filter itself can make things diffuse. I tried filters to improve contrast; but the images lost sharpness and I've never done it again.

Also, to satisfy my curiousity: are you using paper negs or film?

 

[Murray@uptowngallery]

I had same question...film or paper, but with a red filter on paper I imagine he'd be complaining about no image rather than unsharp ones.

With a 25 filter, try about 600-620 nm or so in Pinhole designer. I second the 1.5-1.56 numbers for pinhole constant. These lower numbers are supposed to be better for MTF, contrast and perceived sharpness on larger details (see Carlsson's '...Revenge of the Simple Minded Engineer' paper on Google).

My experience has been that small details in subject matter like landscapes enhance my feeling of lack of sharpness because the details in the smaller formats I shoot are probably too small for pinhole's relatively low resolution.

I tend toward wide angles so very close and large objects work better for me.

For more fun at closer than landscape/infinity distance, look up the Paul Prober/PinPlus site and look at his so-called Prober-Wellman paper for pinhole size at macro-range distances...the hole required gets significantly smaller.

If you don't want to wade thru the math (most don't), lean toward a smaller aperture for closeups...ballpark 1.25 constant.

Only problem with optimizing a camera for a specific distance other than infinity is it becomes uncooperative at infinity when you need it. I made one for 1" and it was lousy at 4 feet. It became a tool for only one job.

Stop by and visit.join f295.org too!

 

[Jim Jones]

Using Pinhole Designer, there is a noticable difference in recommended pinhole diameter between wavelengths near the blue and the red end of the visible spectrum. Long ago I compared resolution tests between blue sensitive film and Kodak IR film with a 25 filter. There was a modest difference in optimum pinhole size, well within the tolerences that many people accept. The gain in sharpness by using the UV end of the spectrum may be more than ofset by the lack of panchromatic film response.

Rich, your neighbor would certainly be right in regard to simple lenses with no color correction. In theory he is right in regard to pinholes. but the gain is very slight in practical pinhole photography.

Quality filters shouldn't degrade sharpness except for any change they make in color.

As Greg says, larger formats can record more subject detail. The blur in a pinhole image is proportional to the pinhole diameter. That diameter increases in proportion to the square root of the focal length. For equal angular coverage, the film size is proportional to the focal length. Thus, a 4x5 pinhole camera can record subject detail about twice af fine as can a 35mm pinhole camera.

 

[Joe VanCleave]

I limit the spectrum of light in my pinhole shooting by using ortho media, such as paper negatives or graphic arts film; these are mainly sensitive to UV and blue. I find this more convenient than starting with faster panchro film and having to use a filter.

I believe larger formats, suitable for contact printing, are key to sharp images, along with a smaller sized pinhole optimized for blue/UV-sensitive ortho media. Of course the downside of the restricted tonal range of ortho media may offset any aesthetic advantages to a sharper image; only you can decide that for yourself.

If you decide to shoot mainly closeup images, such as dioramas or still-lifes, then you can get by with a much smaller pinhole size. At such close-in distances, the image-softening effects from geometry are stronger than the image-softening effects from diffraction. Stated another way, most formulae for optimal pinhole size are themselves optimized for objects near or at infinity. For instance, an object that's 1F in front of the pinhole (i.e. the same distance in front as the film plane is behind) the blurr effect at the film plane from geometry is twice the pinhole diameter, which far outweighs any diffraction effects.

 

[jon koss]

Hi Rich - A couple of ideas for consideration. Not necessarily the gospel though!

Since you have been investing in pinhole for years, it may be worth switching from spending lot of time to spending a little money. If you buy just one commercially manufactured laser-drilled pinhole, then you will have a known "gold-standard" as a reference. The thin wall, blackened circumference and sharp edges of the commercial piece might save years of misplaced effort. If the commercial hole is sharper for you then you are further along the path. If not, then you still have a righteous pinhole for your next camera. Regardless of outcome, you will truly know how good (or lacking) your own pinholes are... and what's that worth?

Next, is it possible that contrast is an issue? Could light be bouncing from your film to the wall of the camera, and then back to the film? Assuming you are painting the inside of the box, some matte black paint can bounce or scatter light more than you might think. Maybe a switch to a black velvet lining inside the camera would help. Or baffling. Or both!

Also, are you totally sure that your box material is 110% opaque? Far out question, but one never knows.

Along the same lines, are you "smoking" the edge of your pinhole aperture? This can help to eliminate stray light that might degrade an image.

Crazy as it sounds, sometimes problems can be diagnosed by "seeing what the film sees." If possible, try getting an eyeball onto the film plane of the camera. This will of course require totally excluding all other light sources, so you may find yourself staggering around the yard with a camera duct-taped to your head! Leave either your nose or mouth uncovered!! In any case, once your eye adjusts you may see obvious problems such as internal reflections, light leaks, hot spots, etc. (This technique also works wonders with Feds and Dianas!)

Same logic in revese applies to film holders. Go into dark room. Install holder, pull slide, remove pinhole, shine flashlight in through front of camera and look for leaks from the outside.

Finally, have you tried a few varieties of film? Say, a controlled experiment with T-grain fast and slow film along with conventional fast and slow film.

It would be instructive to try a controlled test with supposedly undersized or oversized pinholes as well.

All of this will leave you with no time to actually shoot the damn thing, so I promise you will no longer be vexed with unsharp images!

Jon

I have been making phole images for a few years, and have several home-made boxes. I projected all the pinholes on the wall, got an exact measurement of size, did the math to figure out the optimum "focal length," and inspected the pinholes under a microscope for any defects...I'm happy with the images, yet they don't appear as sharp as some others I see...

 

[Greg_E]

Thickness of the metal could be a bigger problem than it is often given. You can buy a sheet of .001 stainless or brass from http://www.smallparts.com that will last you a very long time. While you are there, you can get those tiny little numbered drills, spin them fast in a dremel tool and they make great round holes in the thin material. When drilling, back up the .001 metal with something much thicker of brass or aluminum, this will help prevent the metal from distorting due to the pressure of drilling.

 

[Jim Jones]

My local Ace hardware store has a packet of brass shim stock from .001" up. For larger pinholes .002" is easier to work with. The old method of dimpling and sanding the brass thins it down anyhow.

 

[Rich Ullsmith]

Thanks everybody, for giving me a few things to think about.

I shoot from those old vietnam era 50cal ammo boxes . . .divided in half, each side holds a 4X5 sheet. I made a mess of pinholes in dead-soft stainless steel shimstock, I want to say .01mm. I measured them all, inspected them under a scope, and installed them in the cameras so that two very similar apertures are on each camera, for bracketing. However, I did not smoke the pinholes, the benefits of which I have known for some time, mainly out of laziness.

The cameras are not only light tight, but airtight and water tight as well. The 58mm filter rings are glued on, and the filter over that, then another blacked out filter over that as a lens cap. Light leakage is definitely not a problem.

I'm just saying, I have seen pinhole images, albeit on a computer monitor, that looked to me to be focused light. I know a laser drilled hole would be an improvement, but I've seen mine up close, and there's nothing wrong with them. I know the black will help, though. But buying anything for my boxes, vice scrounging, would rob me of the perverse pleasure I get from this stuff.

 

[Murray@uptowngallery]

I've never smoked an aperture. A few other things, perhaps, decades ago, but I can't recall.

Don't count on a laser-drilled hole being any better. I have some with nasty slag around the aperture. I don't know if they are typical or missed a cleanup process after the laser one. They were gracefully donated to me so I'm not going to badmouth them too much. I do have another size that looks clean.

I tried some loose silicon carbide & water but didn't have the patience to wait for something to change...got a few scratches in the SS after a while.

I may try one of those imitation Dremel bits with the 'diamond' grit under a microscope & see what I can do to get them flat.

Many people have observed better results with homemade ones because some of the homebrew procedures result in thinning of the aperture material.

There are procedures to alter aperture edges called apodization. Comes from the Greek for footless. This refers to the classic undulating Bessel waveform crossing thru zero repeatedly causing the intensity variations with distance. A typical apodized aperture may be altered to remove those variations and produce a single Gaussian curve without all the repeating maxima and minima.

Homebrew holes that result in other-than-perfectly-flat-and-round apertures may be accomplishing some form of asymmetrical apodization.

I want to experiment with non-round holes. I know the math is different and the maxima/minima mentioned above are different, and there can be some ugly artifacts like lenses with weird shaped apertures. I am not sure I believe yet that roundness is important because the square aperture is one of the examples in diffraction theory (slit, square then round in that order I think). I think the different theoretical behavior of non-round holes may throw off comparisons to round when designing with rules for round holes.

If you have the motivation to make your own holes, and are checking them I'd say you may as well keep doing that.

 

[Murray@uptowngallery]

If your apertures are 0.01 mm, they are far smaller than anyone's calculations for optimal at a distance that covers 4x5.

That could be a reason for excessive blur.

 

[Jim Jones]

I can think of two good reasons for non-circular pinholes. Elliptical pinholes should (in theory) give slightly better images in some panorama cameras. Square pinholes can be made adjustable, so one pinhole can be used at a wide range of focal lengths. Perfecftly round pinholes and lens apertures are usually preferred, although my Leitz M-mount Super-Angulon f/3.4 21mm has a nearly square aperture at smaller openings, but the images are certainly good enough.

 

[Rich Ullsmith]

Sorry, Murray, I meant the shimstock I made the holes from are on the order of .01mm, not the aperture itself.

 

[Murray@uptowngallery]

Oh, that's different :O)

 

[Murray@uptowngallery]

If you tend toward building alot of cameras of the same f.l. or want to share them in a group buy, look up emdiasum, a supplier of electron microscopy materials.

They sell electron microscope 'grid' apertures, called 'single hole', or 'Gilder Grids' by some suppliers. They are electrodeposited copper, very thin, with clean round holes in a range of sizes. A little tiny to handle (3 or 3.15 mm diameter), cost about $20-25 per vial of 100. Users find them to be as good as or better than laser-drilled.

They will darken to a dark gray in selenium toner.

I pick them up with black photo tape when ready to mount them. I gave up on tweezers. Maybe those little vacuum pickup tools would work well if it's a habitual thing.

 

[Ole]

If you tend toward building alot of cameras of the same f.l. or want to share them in a group buy, look up emdiasum, a supplier of electron microscopy materials...

That's www.emsdiasum.com

 

[Rich Ullsmith]

Going there now. Thanks.

 

[panastasia]

.. When drilling, back up the .001 metal with something much thicker of brass or aluminum, this will help prevent the metal from distorting due to the pressure of drilling.

How about making a tight sandwich with the thick stock above and below the thin stock to minimize burrs when drilling?

I'm not a pinhole photographer but became interested in making one after seeing the work of Barbara Ess in here book I Am Not This Body. The unsharp images were the best, IMO, very surreal and interesting compositions.

 

[Jim Jones]

How about making a tight sandwich with the thick stock above and below the thin stock to minimize burrs when drilling?

With very thin (like .001") that certainly helps. I also debur pinholes with a needle with the tip honed to a broad three-sided pyramid.