Projects and ponderings for film photographers

Archive for January, 2008

Albert Einstein Invented Auto-exposure!


Reading a recent Einstein biography, I was most amused to learn this little tidbit: In 1936, Einstein and an associate named Gustav Bucky were issued US Patent No. 2,058,562 for a “Light Intensity Self-Adjusting Camera.”

Albert Einstein, Overexposed in 1934

Albert Einstein, overexposed in 1934

The text of Einstein & Bucky’s patent can be seen here, and the accompanying diagram here.

This date is quite early for such radical technology: The earliest auto-exposure camera I know about which actually reached the marketplace was Bell & Howell’s Electric Eye 127, circa 1959. But I’m not sure Einstein’s design was really the first invention of an auto-exposure camera (and I don’t have the patience to research all the patent literature).

If you’re having a little trouble deciphering the patent documents, the basic idea is this: the upper lens aims light from the scene onto a photoelectric cell. This is wired to a meter movement, which deflects through larger angles as the light level increases.

But instead of the usual read-out pointer, the axis of the meter connects to a pie-wedge shape of clear sheet; this has a graduated neutral-density filter along its bottom arc. The arc swings across the light path of the lower, picture-taking lens. Brighter light levels move the denser end of the filter into the light path, keeping the exposure consistent.

You may remember that Einstein himself started out as a Swiss patent examiner. And, I don’t want to be too hard on the father of General Relativity, a Nobel Prize winner, and a certified genius. But there are a couple of problems with this scheme.

Using a neutral-density filter means throwing light away. This would not have been good news in 1936. Kodak’s then-new Kodachrome had a speed of about 8 (expressed in modern ISO terms—the scale did not exist then). So adding a filter would necessitate longer exposures and wider lens apertures—sacrificing depth of field, and increasing aberrations.

Perhaps recognizing the problem, the patent includes a second system of manual aperture stops. These are coupled so that smaller lens apertures also admit less light to the meter cell. However this seems like an afterthought—it brings back complicated manual intervention to a system that was supposed to be automatic.

Another consideration for today’s photo-geeks is whether the gradient filter would do strange things to bokeh—e.g., causing out-of-focus highlights to have an odd “fade-out” across them.

The obvious question to ask is why Einstein and Bucky didn’t make the next logical leap: Directly controlling an iris using the movement of the light meter. This was Bell & Howell’s approach, albeit with a very simplified aperture formed by two crossing, comet-shaped openings. Even then, inertia of the moving parts slowed its response time.

Ultimately the breakthrough for mechanical auto-exposure cameras was the “trapped needle” system, used in countless 1960s and 1970s Japanese 35mm models.

A light meter needle (which could provide an exposure preview in the viewfinder) is pinched and immobilized as the the shutter button button is pressed. The photographer’s finger pressure also closes down the lens aperture from its widest opening; but the needle position limits its travel, halting it at the correct f/stop.

Aside from requiring a long stroke of the shutter release plunger, the system was simple and reliable. But to the best of my knowledge, no Nobel Prizes were ever awarded.


Postscript: A bit more research reveals that the very first auto-exposure still camera ever marketed was Kodak’s “Super Six-20″ from 1938. It’s seen at the bottom of this column by the sadly-just-deceased Burt Keppler. Evidently Kodak sold fewer than 800—so I don’t feel too guilty for being unaware of it. Interestingly Kodak seems to have used a version of the trapped needle design I described above, as discussed in the “Notes” section of this George Eastman House catalog entry.

DIY: Plasticam Pinhole, Part Two

Welcome to Part Two of our project converting a cheap focus-free 35mm “trashcam” into a pinhole camera.

Part One covered choosing the right trashy plastic camera to use, then taking it apart. The crucial point to remember is that your camera must have a little lens-cover flap, which will become the new shutter after the pinhole mod.

Trashing and Hacking

With the camera’s lensboard removed, we expose the original spring-operated shutter blade (see the final photo in Part One).

Pull out the original shutter blade and spring, and throw them away. If you had already removed the lens, throw it away too. But keep its retaining ring—we will need this later.

If the lens was glued into place from the back of the lensboard, you may need to shatter it by tapping a nail through it; then pick out the fragments.

Now you need to decide the best location within the camera’s innards to place a thin metal wafer for your pinhole. The deeper back into the camera body you choose, the wider the angle of view you’ll get in your photos. But if you go too wide, the edges of the opening in the front shell of the camera can creep into the field of view.

With this model I decided to take a chance, and glue the pinhole behind the lensboard. For cameras having only a small opening through the front shell, a pinhole mounted in the front of the lensboard is better.

The original shutter blade pivoted between some molded ridges, both on the main body piece shown here and on the back of the lensboard. With a sharp blade, shave these away, so that there will be clearance for your pinhole’s sheet metal.


The aperture stop in the lensboard also needs to be trimmed away, to insure it doesn’t obstruct the pinhole’s view. But keep the raised collar around the opening intact.

Finally, enlarge the front opening of the retaining ring. This image shows the ring after widening its hole as far as possible.

Remember that our shutter flap will slide against this ring; so its front face needs to remain flat and without rough edges.

(If your lens was glued in from behind so you don’t have a retainer ring, the same consideration applies to the raised collar on the lensboard.)

Pinhole Time

Next we need to choose the right pinhole size, and make the actual pinhole.

The distance from the pinhole to the film plane determines the right diameter for the hole, so find a way to roughly measure this distance on your camera. On my particular model it’s 25mm. This functions like the “focal length” of your pinhole. And whooeee—25mm is a real wide-angle!


You might imagine that the smaller you made your pinhole, the sharper the image would be. But it’s not quite that simple. Light waves grazing the edge of the hole diffract in unwanted directions. So there are formulas for calculating the hole diameter which yields the best sharpness, given a particular focal length.

Actually, using 0.2mm is close enough for most plastic trashcams; but you can get more precise using an online calculator like the one at Mr. Pinhole’s site.

That calculator also tells you the equivalent f/stop of the pinhole, which in my case is f/119. That’s near enough that I’ll round it off to f/128, exactly 6 stops smaller than f/16 (helpful to know when it comes time to determine exposures).

Fabricating a Pinhole

There are several schools of thought on the best method for making a pinhole, so I’ll keep things brief in describing my method.

I tap a small bump into thin sheet metal (pop-can sidewall will work); then sand the bump against 320-grit sandpaper until the metal is paper-thin. I press the bump against something firm like a phone book, and with the very tip of a sewing needle, pierce the tiniest hole I can. The shaft of the needle mustn’t go through.

I use squares of metal 50mm x 50mm—the same size as a 35mm slide mount. This lets you use the slide holder of a film scanner (or, a slide projector) to get an enlarged view of the pinhole. The hole needs to be nicely round, and not ragged; and by knowing the scan resolution (or magnification of a projected slide) you can calculate the diameter.

By gently twirling the needle tip in the hole, and lightly sanding after, you can nudge the diameter larger until you reach your target size. (Blow out any dust before checking it.) You’ll probably need to make a few pinholes to get a good one; but even if the diameter is off by 20%, that’s only a fraction of an f/stop in exposure error.

Place the Pinhole

Trim the edges of the pinhole metal until it fits into the location you chose. If the pinhole is going behind the lensboard (as here), test-fit that there are no remaining ridges or nubs which would keep the lensboard from snapping back into place.


Glue the pinhole in position. I like to use black silicone sealant for this (it’s sold as automotive gasket material), since it blocks light from leaking through any small gaps. Check through the back of the camera that the pinhole is centered, sliding from side to side if needed. Don’t get glue in the hole!

Too Smart For Its Own Good

Next we can start putting the camera back together—with one important final modification.

You may have noticed that your camera originally had a nifty interlock between its lens cover and its shutter button. If the cover flap was closed, the button could not be pressed down.

Once we gutted the original shutter mechanism, the button atop the camera no longer does anything to start exposures (the lens-cover slider will do that instead). But we still need to click the top button each time we want to advance the film to the next frame.


In our finished camera, the cover flap will be closed any time we aren’t exposing film. Hence we need to disable the interlock, so you can click and wind between exposures with the flap still closed. My illustration shows where an arm on the lens-cover slider originally blocked the shutter plunger. Remove the slider, and cut this arm away, and the release button can be clicked even with the flap closed.

Finally, we can put all the parts back together: Snap the lensboard in place, and add the retaining ring. (With the lens missing, the ring may fit loosely, requiring some dots of glue to hold its ears in place.)


Replace the cover flap and its slider, and check that it works smoothly to cap and uncap the pinhole opening.

Time for a Sanity Check

Next, put the front shell of the camera back in place again. (It can take a few attempts to put the front on without dislodging the cover-flap/slider mechanism.) Do not replace the four screws yet—we need to make one important check.

Open the shutter flap; and with the back of the camera open, hold up the pinhole towards a bright light. Sight through the film gate towards the pinhole, rocking the camera back and forth. The pinhole should remain brightly lit from anywhere in the frame—all the way around the perimeter, into all four corners.

Uh Oh!

Trying this test on the camera shown here, I discovered trouble. I realized I’d been too ambitious at trying to get the shortest possible focal length and the widest angle of view. The front shell’s opening blocked the corners of the image!

Remember that a pinhole has effectively infinite depth of field. So the round edges of of the opening would appear sharp in the photos—the pictures would look like they’d been taken through a port-hole. While it might be interesting to play around with that effect, I preferred an unobstructed view.


I solved the problem by taking a file to the edges of the front opening—shaping a bevel with the same rectangular proportions as the 35mm frame. I checked my progress by sighting along the edges of the film gate, and eventually removed enough material: The pinhole had an unobstructed view all the way into the corners. Whew!

With that crisis averted, finally we’re ready to put all the camera parts back together. Replace the latch spring (if yours fell out); the rewind crank (if you removed it); the wrist-strap (if you want it); and the four screws that hold the front shell in place. Make one last check that your shutter flap is opening correctly, and your camera is ready to shoot!

Finishing touches

Because pinhole exposures can be quite long—with indoor light, even many minutes—holding the camera steady during that time becomes a problem.


I suggest adding a flat piece of wood as a base—much easier than the camera’s rounded body to steady against a table, a bench, a door frame etc.

If you have a tripod available, a 1/4″-20 nut epoxied into the base serves perfectly as a tripod socket.

Note that the base needs to be cut short, so it does not cover up the rewind release. (Or if you prefer, drill a large finger-hole lined up with the release button.)


Once you’re certain you won’t need to open up the camera again, glue the base in place (hot-melt glue works great). Just watch out to make sure that the back can still open freely.

Shooting with the Pinhole


This wide-angle style of pinhole camera gives interesting stretched-out, dreamlike images. Both near and far are equally in focus (or equally defocused?); yet anything that moves during the exposure disappears in a ghostly blur. A pinhole camera’s potential for intriguing, expressive images is something I’ll leave it to you to explore.

But if you’re a beginner who would like some basic tips on operating the camera, exposure suggestions, etc., I’ve put them into a PDF file which you can print out and take along when you go out shooting with your new little plastic pal.

Happy pinholing!

[Return to Part One.]

120: The Survivor

I was looking over some creaky old folding cameras recently—probably about 90 years old—and was amused by a curious fact: A couple of them used a film size you can still buy today.

Now, perforated 35mm film has existed ever since it emerged from Thomas Edison’s lab as a movie stock in 1892. (William Dickson is credited with the actual invention.) But only in the 1920s, after the Leica camera, did 35mm gain wide acceptance as a still-camera film.

Even then, bulk cine film had to be re-wound into special film cassettes that were proprietary to particular camera brands. Finally, in 1934 Eastman Kodak introduced the Retina camera, along with a new disposable metal film cassette preloaded with 35mm film. In keeping with Kodak’s other three-digit film designations, the new format was called “135.”

1937 Kodak Retina

My father’s 1937 Retina: 3rd version, three years after the introduction of 135 film

Cleverly, Kodak’s German division (the former August Nagel cameraworks) designed the cassette to be backwards-compatible with Leica and Contax cameras, as well as the new Retina line. Thanks to the popularity of the Leica and the Retina—and perhaps even more so, the affordable Argus—the 135 cassette was a success. Today, it’s the universally-known way to shoot 35mm film (ignoring oddities like 250-exposure backs for motor-drive cameras).

Yet this is not the oldest still-camera format we continue to use today. That distinction belongs to 120 film.

Kodak introduced the “No. 2 Brownie Camera” in 1901, a simple $2.00 box camera designed to shoot images 2-1/4″x 3-1/4″*. And for this camera, Kodak introduced a new film size, 120.

Three Generations of 120 Spools

The earliest spools had wooden cores (this one is from Ansco, who used the designations 4A or B2 for the size), later metal, and finally plastic. But the dimensions of these three spools are the same.

As time went on, 120 was adopted by a wide range of camera styles: From simple snapshooters to the fine Rolleiflex (introduced 1929) and Hasselblad (introduced 1948). Its use in these professional-level cameras helped keep demand for the format strong; so today it remains the most widely-available film size after 35mm.

I need to get on my soapbox for one brief rant: Please, please, please, don’t mistakenly call this size “120mm.” Even B&H’s online store has been known to get this wrong. The film is actually about 63mm wide, and the 120 is just an arbitrary number from the dim mists of Kodak history.

Originally Kodak’s three-digit size designations were supposed to have some meaning: but the rationales were rather obscure, and soon became unworkable as new sizes appeared. In any case, Kodak continued to give new film formats various three-digit monikers, like 127 for “Vest Pocket” camera rollfilm or 828 for “Bantam” size—right up through 110, the drop-in plastic cartridges introduced with the Pocket Instamatic series in 1972.

Today, the roadside of photo history is littered with the carcasses of extinct film formats. The once-common 116 and 616 sizes are gone, and so are 620 and 828. “Vest Pocket” 127 limps along with a single Croatian emulsion still available. Users of 126 Instamatics must scour the internet for the last sources of film.

Yet even at 107 years old, the 120 film format still seems to be going strong.

*Today we might call this “6×9 cm,” although that’s only approximate: the actual negative is more like 5.6 x 8.2 cm.