NOTE POSTED 04-17-2009 – I have been trying to get a handle on the cross slit project since this was originally posted here. At this moment, and I don’t see my mind being changed, I am abandoning any further work or time. I believe that simply put, since there are two distinctly different focal lengths there will always be one plane that is stretched out of proportion. I am planning to continue into the world of old manual focus M42 lenses and even older lenses from folders on my Canon XTi. Of course, there will still be occasional side trips into pinhole land.
Original text starts here:
Here is a “how to” double slit lens assembly project. Please read my warning on the main page regarding who is at fault if you try this and anything at all goes wrong. Hint, it isn’t me. Further, the parts of this lens are right there in close proximity to the internal parts of your DSLR such as the sensor and the mirror. Extreme care should be taken by YOU to ensure nothing will come loose (including paint and glue) and fall inside your DSLR.
This project was undertaken to find out the ins and outs of the project before making a dedicated MF camera. What I made is lens with adjustable slit spacing to go on a DSLR camera for experimentation.
A word or two about the image quality. The small size (most are even sub-35mm) of the sensors on DSLRS have a very hard time with pinhole caps in regards to image quality. 120 film cameras and larger work much better. This project pushes the limits even further – see the pictures below. I am very happy with the very first results- taken handheld with long exposure times on a dreary, rainy Upstate NY Fall day. The ISO on my Canon XTi had to be set up to 1200 in order to get these images. I anticipate that the results will be better after I use a tripod on a sunny day and get a handle on slit width and spacing- both were just wild guesses this time.
Things used: black foam core, Shick single edge injector blades, a set of macro rings, matt black paint, quick set epoxy glue, a single edge razor, an exacto tool, a nest of feeler gauges from an auto supply store and other small things like brushes, etc.
The pictures in the gallery follow the text. I’ll refer to them as 001, 002 etc.
Image 001: Press the camera mount ring and lens mount ring firmly into the foam core. This makes a pattern, cut them out a bit oversize and trim to friction fit the inside of the rings.
Image 002: Test fit the cutouts inside the rings. Cut out openings for the slits. I used a piece of 1/4″ hardboard on edge. Make the openings as close to the center as possible. Glue them in place inside the rings with epoxy glue.
Image 003: Ready for the blades to be placed.
Image 004: Here I have tack glued the razor blades over the openings. I used a .014″/.35mm feeler gauge to make the opening. This approximated a 55mmFL size pinhole opening.
Image 005: The shine from the foam core paper, the epoxy glue and the razor blades have been painted over with matt black paint.
Image 006: All ready to be tested.
Image 007: The area cropped from an a 28mm prime normal lens image for comparison.
Image 008: Taken with just the two mounts joined- notice there is already a distortion in one plane. This distortion is a normal result of this type lens due to the different focal lengths (distance between the sensor/film) of the two slits and is to be expected. The distance between the slits in this case is approximately 7mm. The rear slit is approximately 50mm from the sensor.
Images 009 through 011: Increased distortion as the focal length of the front slit is extended with use of the rings out to approximately 110mm. Remember that in this setup the rear slit is at a focal length of about 50mm.
There is a good possibility I will take another set of rings and mount a rear slit in one of the #1 rings. This will allow the movement of the rearmost slit in addition to the front slit. Then the slits can be both moved in and out and the distance between them adjusted. The way I did this set, only the front slit can be adjusted, the rear slit is always at the same distance from the sensor.
If you want the least distortion – make sure the slits are perpendicular to each other. The multi turn screw sections allow the lens slit to be rotated to the correct aspect, A piece of duck/duct/gaffers tape would hold the in place and allow changes to be easily made.
Future changes will probably follow now that it is apparent that the design of a cross slit lens is viable. Probably it would be changes in focal lengths of the slits and the size of the slit openings. The closer the slits are together, the less the distortion, but how close can they be and still function ? Is there a magic ratio between the slit opening sizes ? Are there slit openings that work best with different focal lengths. The only reference I found said to use the same opening size as would be used for a pinhole for any focal length, is this true ? There is also plenty of room for experimentation with film and paper as the recording medium. I do hope others will join in and have fun answering these questions with actual physical experimentation rather than scientific theory and mathematical formulae.
