[rollei_list] Re: Scanners

Hi guys,
Lots of discussion about scanners but my original question remains un-answered. I had hoped to get a coherent ex plantation from someone on this list who designs scanners or digital cameras but so far just see a lot of argument. Let me repeat my original post in case someone can shed some light on my question. I wrote as follows: 
-----------------------
I think one or two folks on this list actually design scanners and/or digital cameras. I have a background in optics for distance measuring but none in image collection so I ask the following:
How can the depth of field (actually depth of focus maybe) be so large in flat bed scanners that deal with film and transparencies using a background light source in the lid?
I know that if I am simply scanning a flat sheet of text for character recognition, or a photographic print, in the reflected light mode, the sheet or print will go out of focus fairly quickly if not right down against the platen. Books, for example, when the text is close to the binding area, etc. That problem doesn't seem to happen, at least for small distances, with transmitted light scanning.
I have gone through a couple of scanners and am perfectly happy, for my purposes, with the Canoscan D1230U. Works fine for me, even with 35mm slides and negatives, and is great for 6X6.
It comes with several film and slide holders but they all seem to put the film several mm above the glass. Apparently that doesn't really affect the sharpness of the scanning.
Just for the heck of it I made a couple of holders which work with film and un-mounted slides, which put the material right down on the platen. Doesn't seem to make any improvement.
Should I assume that the light-collecting diodes are really very narrow beam (acceptance angle) devices or is something else in play here? 

The above may not be well-posed but I think you can understand my question.

Thanks,

DAW

Other related posts: