All images are out of focus with pinhole body cap

Question

I bought a "Rising" brand pinhole lens. It's a piece of drilled aluminum fixed to a body cap, the kind of plastic cap that covers the body when there's no lens attached.

I have a Pentax K-50, and I got the appropriate lens. The first thing that happened when I turned the camera on was that it asked for the focal length. After learning what that was, I figured it made sense-- no electronics in the cap to communicate such a value.

The lens came with a paper indicating focal lengths for about 8 common manufacturers and a representative model. For Pentax, they listed a focal length of 48.5 for the 135-K.

My camera didn't allow such a precise value-- the closest values were 45 and 50. I tried them both, but the images were out of focus.

I googled for more information, but didn't find any. Frustrated, I shot images at all the focal length settings, from 8 mm to 800 mm. The UI is using one of the jog wheel to navigate through preset values.

I ended up with 35 images, all out of focus, and all appearing to be the same. Whatever the correct focal length is for my camera, I don't seem to be able to set it.

I'm getting ready to return the item, but I just want to put the question out there, is there anything else that I am overlooking? I emailed the manufacturer about the setting for my camera a few days before my comprehensive test, but I haven't heard back from them.

I used a tripod and electronic "bulb" actuator. The exposure is good; I nailed it down with the method of halves.

My thought is to take the money I spend on the lens, and buy some extra lens caps, and drill various size pinholes in them.

Is there anything that I've done wrong, or will this lens simply not work with my model camera?

Edit The diameter of the pinhole is 0.22 mm.

This is one of my test photos with the pinhole:

And here's a shot with a 50 mm of the same area:

Maybe I'm beating a dead horse, but here's a comparison of the printing on the PVC pipe:

I recall reading that pinhole lenses allowed for almost infinite depth of field. I googled some images, and while some have motion blur (at times a lot) and also a big edge distortion, the center area seems to be pretty well focused:

If my tests the expected result from a pinhole lens, I wonder, what is the point of it? How do I get the sharp images one finds in a google image search of "pinhole"?

Answer 1

This is exactly what one expects from a pinhole lens.

Normal lenses

A normal lens has a big opening and bends all the light rays coming from a point-like object to cross in a point behind the lens. By moving the lens fore and back, one can determine the position of the point behind the lens, and when it's exactly on the sensor, it might hit a single pixel only. (In theory.) So, a point-like object would be projected to a single pixel, which is the maximum sharpness you can get.

At the same time, light rays from other point-like objects in other distances are bent so that they cross at a point in front of or behind the sensor. This gives circles of light on the sensor, hitting many pixels. So, those objects are unsharp.

This is illustrated in the following image I made using a simulation. There is a 50mm lens and three point-like light sources placed 800, 1000 and 1200 in front of the lens. A screen (=sensor) is placed 52.63mm behind the screen, where the light rays of the middle, red light source cross.

(Click to enlarge)

This zoomed-in image shows how the blue and green rays cross in front of / behind the screen and form larger spots on the screen. (The black dot is the focal point of the lens)

Pinhole lenses

A pinhole lens only has that pin hole, but no optical lens. That means that the rays entering through the pinhole will not be bent to cross at a point. They travel in the same direction as before, and form a spot of about the size and shape of the pinhole on the sensor. This means, if your pinhole has 0.22mm diameter, you get a 0.22mm circle on your sensor. For example, my camera has a 22mm wide sensor and the photos are 5200 pixels wide. The 0.22mm circle-shaped spots from the pinhole lens would result in circles of 52 pixels diameter on my photos. Quite blurry!

Now, while the optical lens inside a normal lens can move forth and back to focus objects in different distances, the rays from objects in all distances entering a pinhole are about parallel. (Except for objects very close to the camera) So, all objects show about the same unsharpness.

Here are images with the optical lens replaced by a pinhole. Since a 0.22mm pinhole would not be practical here, I exaggerate it by using a 5mm pinhole, but the principle is clear:

Due to the heavy overlap, the three light sources could not be distinguished.

Setting the focal length

I wrote that focal length is a pure informational value in your case, but as @junkyardsparkle and @mattdm pointed out, that's not true.

Today, cameras have image stabilization techniques to reduce blurriness due to small shivering movements when the camera is hold in the hands. Some cameras like DSLRs from Canon move the lenses sidewards, while others like yours can move the sensor instead. The focal length is needed to calculate how much the sensor has to move to cancel out a detected shiver.

Diffraction?

An other user wrote in his answer that diffraction is the cause of the blurriness of the image. Well, it's not that easy here.

In general, diffraction exists, and I'd like to link to Wikipedia about Airy disks for further details.

The important formula is

• λ is the wavelength, for the visible light 400-670nm
• d is the diameter of the pinhole
• θ is the angle between the center of the airy disk and the first ring of absolute darkness around it (which can be seen as boundary of the disk), as seen from the center of the pinhole.

Now, with your numbers, one gets θ=0.001818rad for violet 400nm light, and θ=0.00302rad for red light. This multiplied with 48mm, the distance between pinhole and screen, gives the radius(!) of the airy disk, so one has to multiply by 2 to get the diameter. Then, the diameter is 0.17 to 0.29mm.

OK, so indeed, the spot on the screen is larger for red light if diffraction is considered. But wait, it's smaller for blue light?

Well, those formula imply that the diameter of the spot is zero directly behind the pinhole, which obviously is not the case. Those formula are correct far, far away from the pinhole, and our screen is not yet so far away.

This image demonstrates this. About 100mm behind the pinhole, the spots are really much larger due to diffraction, but at about 50mm, well, a little.

In case of a normal lens, diffraction plays a bigger role at high F-stops. A perfect lens would focus the light beam to a single point, but the diffraction makes it a spot with some diameter, or even some pattern.

Notes

One could write a book about everything that affects the image quality. For example, the real lenses here are considered ideal. For the pinhole lens, the precision of the hole, but also the thickness of the material plays a big role.

Answer 2

Your results are GREAT. You show exactly what one should expect using a "pinhole" lens.

The reason your results are different from your Google search is because the image that you found wasn't produced by a typical pinhole lens in a single typical session without some manipulation in printing. You chose a photo from a series of creative pinhole photography "departures." Some in the series were made with multiple negative exposures in printing, some with negatives that were simple photograms. It would be difficult to know exactly how much of the work is with a pinhole lens exclusively with a straight print.

Now what? Try pushing the "unlimited depth of field" to the limits. (Don't confuse it with things like resolution, sharpness, flare, and things that have to do with lens formulas for glass.)

Try setting up a situation that you know would be impossible with a lens. Move your camera a few inches from a textured wall and include part of the wall with something in the mid-field such as a car, and mountains in the horizon and you'll see that they are all equally imaged.

You've already discovered the soft image quality. How do high-contrast scenes look? What images would be complimented by such a specialty/novelty image treatment. Usually, one doesn't crop a pinhole-lens image. Enjoy!

Answer 3

Images you see done with pinhole online may be sharper for several reasons.
Size of the original... 4x5" film
Quality of the pinhole.... laser cut vs drilled vs pin pushed through foil
flare from light source
color vs b&w, monochromatic with filtration will be sharper

Otherwise you are on track. You should buy a book on pinhole photography, to learn how to best control and create what you wish. Part of the fun is not having control. On the other hand it's also nice to have a choice of 6 calibrated precision drilled pinholes to have control and options. https://www.pinholeresource.com/ has some great books and equipment.

Answer 4

Pentax K-50 has APS-C sized sensor (23.7x15.7mm). Flickr photographs mentioned here taken with 120 type film with frame size 90x60mm - area ~14 times larger than K-50 sensor therefore better apparent resolution. Sometimes people shoot with even bigger sheet film. Pinhole photo resolution is limited physically by diffraction (you cannot decrease hole diameter further: for 50mm focal length optimal is around 0.24, 100mm — 0.35mm, for 24mm — hole diameter would be 0.17mm, however you cannot fit 24mm with DSLR - mirror would be on the way). Your sample is quite good as for APS-C. Best way to increase pinhole photograph resolution is to use photographic material with greater area. There is almost no point in using pinhole for APS-C and even Full Frame digital sensors unless you are ok with blurry pictures. If the film is too expensive for you, try photographic paper.