Uexpectedly blurry images on telephoto pinhole

gattu marrudu · Apr 27, 2012

Hi there,
I have built many pinhole cameras in the latest 6 years, and I am now building my first telephoto one.
I basically cut a 2" PVC pipe, lined it with black velvet and glued a 35mm body cap on one end, while on the other end I put a pipe fitting with a an aluminum foil with a pinhole - that way I can change pinhole size easily. Internal reflection is completely absorbed by the velvet, and the tonal range is really good .
I made a 1.2mm pinhole for a 250mm tube, which is about 1.7 times larger than the optimal size recommended by Pinhole Designer and other manuals (0.7mm).
The results are very, very blurry, so much that I can barely distinguish my subjects.
I have made pinhole cameras with much larger holes compared to their optimal sizes, and they never looked so fuzzy.

Logically reasoning I thought that this might be due to the narrow FOV.
If I used a, let's say 4x5" film instead of a 35mm, while keeping the same hole diameter and distance from the film, the circle of confusion would have the same size but on a larger image, thus making the image appear less fuzzy. I could then say that a narrow FOV will give a fuzzier image to begin with, even with an optimal aperture.

I haven't built an optimal pinhole for this length yet, but I don't think it will be useful to my project. I might just go ahead and try just for testing purposes.

Has anybody experienced a similar issue?
gm

DWThomas · Apr 27, 2012

I have not tried a telephoto pinhole, but I think your analysis is going in the right direction. You are effectively looking at a "crop" of what would normally be a much larger negative.

Pinhole Designer shows a 0.7 mm hole at 250 mm, using the constant of 1.9, but gets even smaller at a constant of 1.6 or similar values some people use. Going the other way, using the 1.9 constant and a 1.2 mm hole shows the optimum focal length to be about 960 mm which would be almost four times what you have -- probably ideal for an 8x10 or 11x14 (inch) monster camera! The optimum sized hole would help some, but I suspect it will still be disappointing.

John Austin · Apr 27, 2012

I have been shown the mathematics of why long FL pinholes are problematic - I can't remember the figures, but they made sense - Diffraction becomes a big issue unless a very thin material us used for the hole, or something

Anyhow, try thinnest possible pinhole material and be prepared to accept "interesting" definition

Anyway it is only pinhole nuts like my wife Rae who is on a chase for sharpness in her current pinhole work, spending my hard earned money on micro precision drilled pinholes from Pinhole Resource - These are very good and I have one on a Speed Graphic - NO, I am not going to be a pinholographer for more than one day a year, on Sunday - Up super early and out in the dawn light to get Mondo Numero Uno again with the darkroom pre-warmed, the scanner ready to scan and a dish of metho to dry the film

andrew.roos · Apr 27, 2012

I'm not a pinholer, but as an engineer I think simple optics is the problem.

With a 1.2mm pinhole diameter, even a subject at infinity will only resolve to a 1.2mm diameter circle on film. This gives you a resolution of about 30 x 24 on 135 film. Closer subjects will reduce resolution further. In order to improve resolution, you need to reduce the diameter of the pinhole; but at some point you will run into problems with diffraction, since the long focal length will magnify the ray divergence due to diffraction.

[edit] OK, I did the maths and I think you will become diffraction limited with a pinhole diameter of around 0.4 mm. At this point, the pinhole diameter is the same as the airy disk diameter and the total uncertainty (calculated as the RMS of the airy disk and pinhole diameters) is minimum at 0.55mm. That still only gives you resolution of 66 x 44 or so on 135.

Calculations were for green light (500 nm) but blue light doesn't gain you much - it only reduces the RMS circle of confusion to 0.52mm.

So the problem with long focal lengths is that the f-number is small (f/625 in our example with a 0.4mm pinhole) which puts a lower limit on pinhole size (due to diffraction) that is still too large for good resolution (due to geometric optics). This is a fundamental physical problem, nothing you can do about it (except to reduce your pinhole size to 0.4mm to get the best possible resolution at that focal length).

polyglot · Apr 27, 2012

"Optimum" diameter is where the pinhole size is the same as the circle of confusion from diffraction; that size is the output resolution you'll get.

For tele pinhole, there is more diffraction so less resolution and you can make the hole larger. You made the pinhole larger again, reducing the resolution even further. Note that you will never get resolution finer than your hole size.

gattu marrudu · Apr 27, 2012

I'm not using the most high-end material for my pinhole, that's for sure - aluminum from a soda can with a hole either drilled while clamping the foil between two thicker aluminum pieces (for holes >1mm) or the old pin-and-sandpaper method (<1mm).

I don't know how much a fine edge will affect a hole that size though.

@Andrew, you brought up light color that might add further to the fuzziness (I didn't know about that). I have tested on night street scenes, lit with mercury lamps, thus heavily biased toward the red end of the spectrum.

So, as far as I understand the optimal pinhole radius is the point at which the diffraction and the circle of confusion balance each other. If the distance from pinhole to film increases, light rays diverge further and the diffraction effect becomes more apparent, and you need to open up the pinhole to compensate that. But if you keep the same frame size, the COC becomes larger on the film.
So, apparently the "optimal" pinhole size for a 10° FOV is less "optimal" than the one for a 50° FOV - correct?

Anyway, here are a couple of shots I took using a digital SLR - which means the FOV is even narrower, comparable to a 320mm for a film camera. As you can see, the results are pretty unusable. Pinhole sizes are 1.2mm and 1.8mm (the one I would like to use for exposure purposes).

andrew.roos · Apr 27, 2012

gattu marrudu said:
@Andrew, you brought up light color that might add further to the fuzziness (I didn't know about that). I have tested on night street scenes, lit with mercury lamps, thus heavily biased toward the red end of the spectrum.

So, as far as I understand the optimal pinhole radius is the point at which the diffraction and the circle of confusion balance each other. If the distance from pinhole to film increases, light rays diverge further and the diffraction effect becomes more apparent, and you need to open up the pinhole to compensate that. But if you keep the same frame size, the COC becomes larger on the film.
So, apparently the "optimal" pinhole size for a 10° FOV is less "optimal" than the one for a 50° FOV - correct?

Red light has a longer wavelength (around 700 nm) and so is more affected by diffration than blue light (around 460 nm). Theoretically you could get about 50% more resolution if you used only blue light, and optimised the pinhole for the diffraction of blue light. In practise, I doubt it would be worthwhile.

The size of the pin-hole determines the angle at which the light will diverge due to diffraction. The smaller the pinhole, the greater the angle. However when your FOV is 10 degrees, then a 0.1 degree divergence represents 1% of your FOV. The same pinhole on a 50 degree FOV camera would still cause the light to diverge by 0.1 degree, but now it's only 0.2% of your FOV. This is why diffraction is more of a problem with a narrow FOV. So, as you put it, the optimum pinhole is less optimal on a narrow FOV than on a wide FOV camera.

ME Super · Apr 27, 2012

andrew.roos said:
I'm not a pinholer, but as an engineer I think simple optics is the problem.

With a 1.2mm pinhole diameter, even a subject at infinity will only resolve to a 1.2mm diameter circle on film. This gives you a resolution of about 30 x 24 on 135 film. Closer subjects will reduce resolution further. In order to improve resolution, you need to reduce the diameter of the pinhole; but at some point you will run into problems with diffraction, since the long focal length will magnify the ray divergence due to diffraction.

[edit] OK, I did the maths and I think you will become diffraction limited with a pinhole diameter of around 0.4 mm. At this point, the pinhole diameter is the same as the airy disk diameter and the total uncertainty (calculated as the RMS of the airy disk and pinhole diameters) is minimum at 0.55mm. That still only gives you resolution of 66 x 44 or so on 135.

Calculations were for green light (500 nm) but blue light doesn't gain you much - it only reduces the RMS circle of confusion to 0.52mm.

So the problem with long focal lengths is that the f-number is small (f/625 in our example with a 0.4mm pinhole) which puts a lower limit on pinhole size (due to diffraction) that is still too large for good resolution (due to geometric optics). This is a fundamental physical problem, nothing you can do about it (except to reduce your pinhole size to 0.4mm to get the best possible resolution at that focal length).

Yoda says "Obvious the solution is. Bigger film you must use." Many years ago when I was in High School I built a 4x5 pinhole camera for shop class. A couple years ago I built a 35mm populist pinhole camera. I got better results with the 4x5 than I did with the 35mm. I think andrew.roos calculations are valid. If the optimum pinhole diameter is 0.4mm and you get 66x44 or so on 135, then by enlarging it 4x to get a 4x6 print your COC is far too large at 1.6mm. If you're using 120 in 6x9 format your resolution with a 0.4mm pinhole becomes roughly 150x225. Enlarging to a 4x6 print gets you a magnification factor of 1.69, so your 0.4mm resolution on the film gives you .68mm on the print. You'll get even better resolution on the 4x5 with a 0.4mm pinhole (something like 254x317), and since the film is so big you just do a contact print, which will look even better than 6x9 on 120, and completely blows 35mm out of the water.

Nicholas Lindan · Apr 27, 2012

The sharpest pin-hole photographs are made using the largest film the pinhole will illuminate.

At 250mm (10") focal length you should be thinking of 11x14 or 16x20 film - not 35mm. Contact print your 35mm negative and stand back from it as if it were a 16x20 print and see if it seems sharp now...

'Telephoto pinhole' is a bit of an oxymoron.

Tom Miller · Apr 27, 2012

Grumpy Old Man said:
Anyway it is only pinhole nuts like my wife Rae who is on a chase for sharpness in her current pinhole work, spending my hard earned money on micro precision drilled pinholes from Pinhole Resource - These are very good and I have one on a Speed Graphic

Try electron microscope slides for apertures. They are precision holes in thin metal. If you buy direct from the manufacturer, they're about $25 for a vial of 100. I have twenty of them on one camera, and have seventy or eighty other single-shot cameras with electron microscope slide apertures. They're fantastic, especially when a bunch of cameras need to act the same and not be subject to the variability of hand-made pinholes.

Earl Johnson sells them mounted for $1 each plus S & H.
http://www.f295.org/Pinholeforum/forum/Blah.pl?m-1283887930/

gattu marrudu · Apr 28, 2012

Tom Miller said:
Try electron microscope slides for apertures. They are precision holes in thin metal. If you buy direct from the manufacturer, they're about $25 for a vial of 100. I have twenty of them on one camera, and have seventy or eighty other single-shot cameras with electron microscope slide apertures. They're fantastic, especially when a bunch of cameras need to act the same and not be subject to the variability of hand-made pinholes.

Earl Johnson sells them mounted for $1 each plus S & H.
http://www.f295.org/Pinholeforum/forum/Blah.pl?m-1283887930/

Your URL is broken (or the forum site is having problems), but that could be a very interesting source. I will research further.

I just wonder how much a stainless steel laser drilled pinhole vs. a soda can & sandpaper hole makes a difference with apertures above 1mm.
I'm not looking for a sharp image, I just need enough resolution to capture some details in my subject.
I wondered why there are so many wide angle and so few narrow angle pinhole cameras out there!

Regarding the COC/FOV factor, it would be very useful to be able to predict the resolution of an image.
Can I say, as a general rule, that the COC is dependent only on the pinhole size, regardless of the distance from the film (leaving diffraction effects aside) - or is there some relationship with distance too?

andrew.roos · Apr 28, 2012

Excluding diffraction, the CoC for a subject at infinity is the same size as the pinhole.

The reason for this is simple. For a subject at infinity, all light rays from a point on the subject are parallel. The rays that pass through the pinhole keep on going without converging or diverging (because they are parallel) until they hit the film. So they area they expose on the film is the same size and shape as the pinhole.

Tom Miller · Apr 28, 2012

gattu marrudu said:
Your URL is broken (or the forum site is having problems), but that could be a very interesting source. I will research further.

Gattu, I apologize. The link works if you are a member of f295. I think there may now a cost to get access to the
f295 website. Send me a private message.

Tom Miller · May 2, 2012

If you can't get into f295... here's Earl's email address.
earl.johnson@theobscurecamera.com

RalphLambrecht · May 3, 2012

what is your focallength,distance between pinhole and film?

gattu marrudu · May 3, 2012

I made three different sets - 150, 200 and 250mm on a 35mm camera (taking the distance between the lens mount and the film into account as well, of course). I am testing on the 250 with a 1.2mm pinhole.
I got some .001mm stainless steel shim and I'll see if the pictures look a bit sharper with a more precise pinhole.

polyglot · May 3, 2012

A 1.2mm hole means you're going to have a resolution no better than 20x30. It's going to be a spectacularly blurry mush.

gattu marrudu · May 4, 2012

polyglot said:
A 1.2mm hole means you're going to have a resolution no better than 20x30. It's going to be a spectacularly blurry mush.

"Spectacular" is the right word...

I shot at 3 different FOV's and compared a detail resized to roughly match in size in all the three pictures (see attached image below). Actually, the one that has the best resolution, with the same exposure (I did not correct the tones in the digital image), is the longest tube (the top frame).

This is a digital shot of course, shot at 1600. Strangely enough, with a Delta 3200 exposed for the same time and developed at 3200 I keep getting blank frames, even taking the film reciprocity failure into account.

EDIT - sorry, the last line shouldread 150mm - it's a quick and dirty shot (and of course way underexposed)

Uexpectedly blurry images on telephoto pinhole

Advertiser

Privacy & Transparency

Privacy & Transparency