View Full Version : Pinhole Sizes

06-11-2005, 11:20 AM
Ok, I'm curious about something. Is it prefereable to use the smallest pinhole possible that is wider than the material it is made from is thick? Or to use one of the "calculators" available to calculate the optimal diameter/focal length? If my pinhole is smaller than the size I got from the calculator, will I get a sharper image?


06-11-2005, 02:01 PM
The smaller the pinhole, the more sharpness you lose to diffraction. The bigger the pinhole, the more sharpness you lose to simple geometry -- the pinhole size puts a lower limit on the circle of confusion, which is a measure of the blur in the image. Making the hole smaller also requires longer exposures, which can degrade image quality via the contrast increase from reciprocity failure as well as from increased subject or camera movement over longer times.

The "optimum" pinhole size (for objects more than about 5x the focal distance beyond the hole -- closer objects can gain from a smaller hole) is determined by the tradeoff between geometric sharpness and diffraction. I use SQR(FL/25), in millimeters, which I got from one of the various articles on the web.

There is no optimum for thickness -- the thinner the material, the better. Effectively, you have *two* pinholes, one at the front surface and one at the back, and the closer together they are, the less loss there is to double diffraction; also, the shorter the "tunnel" part of the hole, the less internal reflection affects the image. Ideally, the hole would have a knife edge all around, but in practice the .001" brass shim stock, or similarly thin aluminum can material, carefully sanded to remove burrs, seems to be thin enough, since we're usually using holes bigger than .010" diameter anyway.

06-11-2005, 02:14 PM
Thank you, that's very helpful! I typically use bits of soda can for my pinholes. I've have access to a drill used for printed circuit board assembly that I use to drill my holes. It's given me very clean results. It seems to me that a lot of the pictures I see posted here are sharper than mine and I thought perhaps a smaller hole would work better.

So if my focal length is 110mm, SQR(110/25) would be my optimal diameter? I'm slightly confused by that, the result is about 2.1 if I'm doing it right. That couldn't be in mm's.

06-11-2005, 02:30 PM
Hmm. I think I got the parentheses out of kilter. That should be SQR(FL)/25, so for you, SQR(110)/25 = 10.5/25 = 0.42 mm or .017" hole diameter. That's very close to what I found for my 135 mm FL in my plate camera (not surprising, only about 20% difference in the focal length, or less than 10% difference in the hole size). I think this is about a #78 or #79 drill, if you have tha size -- there are drill charts all over the web you can check against. The other thing is, though, if you don't get all the burrs sanded off (and use fine enough paper to do it) the burrs will add diffraction (= fuzziness). My .018" pinhole isn't as sharp as it should be either; it's nice and round, but I had only 600 grit sandpaper, and you really want 1200 or finer for the final sanding. You can get extremely fine grit "sanding film" sold for making plastic models; I plan to pick some up next time I'm close to Hobby Lobby.

06-11-2005, 02:42 PM
I think we used a .018" bit because we didn't have a .017" when I made this particular pinhole. I'll make a new one and try this sanding film. We'll see if this improves my results.

Thanks again!

06-11-2005, 04:00 PM
I wouldn't worry about .001" difference in hole diameter, but it's worth trying the finer sanding film. Also, if you haven't, you could dry dimpling the metal around the hole and sanding down the dimple to further thin the hole's edge -- but commercial, laser-drilled holes don't do that and always produce excellent results, so I'm not sure how much it really matters.