Close-ups with a rangefinder
Fixed-lens rangefinder cameras aren't really made for macro or close-up work. Most, as is the case with my various Olympus RFs, have a lens in the normal range (40-55mm focal length) and their closest focus is about 1 meter. Compare that to a normal lens on a 35mm SLR which typically can focus down to about 1/2 meter, and there's a difference of an order of magnitude.
As I told one friend, "if it's hard to do I figured I should try". Another interpretation of that, of course is, "Sheesh, why not use an SLR ... you've only got SEVERAL ... not to mention bellows and lenses more suited to macro." Like I said...
Well, the results were pretty pleasing. A slower speed film would have been somewhat better, IMO, but Fuji NPH 400 Professional did really well.
How I did it...
I used an Olympus MCON 40, which is a macro/close-up lens intended for use on some of Olympus' digital cameras. I attached it to the 35 SP by using a 49-55 mm step-up ring; the 35 SP's diameter for threaded attachments is 49mm, and the MCON 40 has a 55mm filter thread.
With the camera empty of film and the MCON 40 mounted, I opened the back of the camera. I had cut a piece of frosted acetate into a 24x36mm rectangle (actually, a tiny bit larger), which served as a type of "ground glass" for focusing. The frosted side was mounted toward the lens, i.e. the smooth side facing the back of the camera.
Moving the shutter speed dial to the 'B' setting and the aperture to f1.7, the lens set at closest focus and in bright sunlight, I moved the camera forward and backward until an object came in sharp focus. Upon achieving focus, I measured the distance from the film plane to the object with a ruler.
The above procedure gave me the correct focus distance for this particular camera. It's not an easy process, but once determined, I can use the ruler to set camera/subject distance with this camera/close-up lens combination. This does not compensate for parallax, which is the phenomenon of the image not be centred in the lens' field of view because the rangefinder mechanism is off centre from the lens axis. (Parallax is only of concern at close focusing distances.) To compensate fo parallax, I simply approximate the offset by eyeballing it. To calculate the actual offset would require more exacting experiments, which I'm not really interested in doing.