Large Format Pinhole Cameras

Building your own camera.

Many people built a pinhole camera while in grade school or maybe as late as high school. It's a novelty, and the results reflect that.

I built one using the carboard box a pack of 35mm film came in, cut out the end and put a piece of aluminum foil with a hole punched in. It produced an image, if you squinted just right, and the results reflected that. The project was never duplicated.

Until last week.

There are many forums dedicated to pinhole photography. I never looked into them. However, I was reading some text books I got on Amazon ($0.01 each, $3.99 shipping, brand new 1000 page physics texts) about light and came across a few things. This got me thinking.

Seems there are certain ratios, and when followed, give reasonable focus to light when followed. In the camera world, there is a thing known as an f stop, or the amount of light allowed in by a lens.  The f stop is determined by the diameter of the lens and the distance to the film plane. The shorter the distance, the wider angle the picture.

The formula for best pinhole size vs. focal length is:

Pd = sqrt(FL)/28

Where Pd = pinhole diameter in mm
FL = Focal Length in mm

Example: for a Focal Length of 125mm the pinhole dia. is sqrt(125) / 28 => 11.18/28 = 0.399 mm

f 300 seems a good size when the distance (focal length) is 125mm.

I use a 4x5" piece of film. 4x5 is a very common size used in large format photography. To hold the film, I got a film holder ($20 on up on Amazon, $5 on up on KEH.com. A 4x5 piece of film, when measured across the diagonal is 6.4", or 162mm.
[square root of (4*4 + 5*5)=6.4", * 25.4 = 162mm.]

A normal lens (ie a 35mm camera has a normal lens of 50mm) for 4x5 is 150-170mm. 125mm is slightly wide angle.

So, with a focal length of 125mm (distance of the film plane to the center of the lens, in this case, the pinhole) we calculate the pinhole diameter. The formula for focal length (simplified) is:

FL = lens diameter * f stop

So, lens diameter = FL/f stop

125mm (focal length) / 305 (f stop) = 0.409mm

My pinhole diameter came out to 0.404mm. Close enough. New calculated f stop is f309.

I took a straight pin, cut off the top part and mounted the rest in a chuck and tightened it down, leaving about 2mm sticking out the end.

(click on the pictures to enlarge)


I placed a 1" round piece of aluminum can (measured 0.002", the thinner the better) on a piece of hard plastic and "drilled" a hole into the metal. At this point, the hole will be ragged. Use some 320 grit sandpaper or finer and sand both sides of the hole. Blow it out, gently poke the head of a pin in the hole to remove grit, blow it out again and check for size.

I placed the metal piece onto my flatbed scanner, scanned it and put the image in photoshop. Photoshop has a measuring tool (ruler tool). I used that and got .3mm. Also, the hole was ragged. I took a needle and very gently rolled the tip in the hole. After a few passes, I got a nice clean hole at 0.404mm.

Actually, I did this half a dozen times until I got what I wanted. Pieces cut from a Coke can are cheap.

I then glued the metal piece with a hole in it to a flat washer with epoxy. While it was setting I made a shutter.

I took some 22 guage aluminum sheet, formed it into a square and folded the edges so that it would hold another flat sheet, slightly smaller. I glued a hex standoff to it, drilled a large hole in it, and epoxied the washer/pinhole lens behind it. This is my shutter.


I used a 1/4" black foam core board I got at Hobby Lobby and made a box to fit around my film holder. I placed my shutter/lens assembly on the center of the front panel after cutting a hole for the lens. In this case, the distance of the pinhole lens to the film plane is 125mm (measured). Plus/minus a couple mm won't matter.

I spray painted the inside flat black and built another box using the black foam core board around the first to cover the seams and make it heaver duty. Since I didn't have any 4x5 film yet, I used a piece of photo paper and, using an ISO of 10, found the daylight exposure of 35 seconds. I exposed the paper for 35 seconds, then removed it and developed it in Dektol. This is the result.





Good. No light leaks, and it actually looks kinda sharp.


Looks like it's worth getting the film. Ordered the film (50 sheets Arista.edu 4x5 film, ISO 100) on Amazon. Arrived in 4 days.

I got some plywood (1/16" thick) and built another box around the foam core box. Then stained it. This made it sturdier and pretty.

The film came in. Too gloomy to take a picture. Stained it again. 12 hours to dry.


Next day, exposure was f11 around noon. Took 4 shots to get the exposure right.

With an indicated f stop of f11, and an actual f stop of f309, I needed to determine the exposure time. I got the indicated exposure time of 1/100 second at f11 using ISO100 film from my light meter.

Another formula:

true exposure = ( ( pinhole f stop / indicated f stop ) ^2 ) * exposure time at indicated f stop

(309/11)= 28 Square this: 28^2 = 789
Multiply this by the indicated exposure time of 1/100 second (0.01 second) = 7.89 seconds. I round this up to 8 seconds.

Using ISO 100 film, here are some example exposure times. I wrote this on a piece of paper I taped on the back of the camera:

Using an ISO of 100:
f16 = (309/16)^2 * 1/100 second => 3.72 seconds, round to 4. Add reciprocity: 8 seconds
f11 = (309/11)^2 * 1/100 second => 7.89, round to 8, double for reciprocity to 15 seconds
f8 = (309/8)^2 * 1/100 sec => 15 sec, double for reciprocity to 30 seconds
f5.6 = (309/5.6)^2 = 3044, * 1/100 = 30 seconds, double for reciprocity = 60 seconds.

I think f5.6 is the slowest I'd shoot. Wind etc. can move the tripod mounted camera and make the whole image blurry. Fine if you want that. 

 Took the picture. My light meter says use f11 at 1/100 second with ISO 100 film. Using the above formula, the corrected time at f309 is 8 seconds. Film has a reciprocity value. When film is exposed at more than 1 second, the silver particles react slower. This means the film needs to be exposed longer. I doubled the exposure time to 15 seconds. Film manufactures websites have the reciprocity values. Arista film is fomopan, and has odd reciprocity values. It's also very contrasty, and with longer development times can be easily pushed requiring shorter exposures. And it's cheap - $35 for 50 sheets. Verses $45 for 10 sheets of Kodak.

I developed the film in a tray in my darkroom for 8 minutes in Kodak D-76. This is the result.



Impressive. Why do I keep buying expensive lenses if a pinhole does this? Anyway, I'm pleased. Not bad for $18 worth of parts (most of the cost was polyurathane and stain).

An interesting effect of long term exposures is that moving things don't show up. If you take a picture of a busy street with cars between you and the object you are shooting, the cars won't show up. Or they appear as ghosts as they pass by. Same with people walking by. The longer the exposure, the more pronounced the effect.

I also made a time chart using photo paper. Photo paper is made for long exposures under an enlarger, and has no reciprocity to worry about.

My f stop: 309. ISO of paper: 10. Going from an ISO of 100 to 12 is 3 stops. (100 -> 50 -> 25 -> 12). ISO of 10 is close enough.

f309/f16 ^2 = 373  exposure time of 1/100 down 3 stops is 1/100 -> 1/50 -> 1/25 -> 1/10

So, 373 * 1/10 = 373 * 0.1 = 37.3 seconds. 40 seconds is fine.
Now we can work out the rest:

f309/f11 ^2 * .1 = 80 seconds (rounded)
f309/f8 ^2 * .1 = 150 seconds
f309/f5.6 ^2 * .1 = 300 seconds (rounded)

I probably wouldn't go below f8.

I think a film rated at ISO200 or ISO400 would be a better choice. Shorter exposure times with film has less of a reciprocity effect. I tried a 150 second exposure with Arista ISO100 film and the results were ugly. This was at f8. My next exposure time was 25 seconds and it looked pretty good. Nice contrast, no weird light issues.


I'm quite pleased with how this turned out. Well worth my time.

Kurt

Comments

Popular posts from this blog

nano80 - An Arduino nano based 8080 Simulator

Our Universe is a Simulation

Lightwave Communications