Determining Scanner Film Height

[Michael Heald]

Hello! What is the best method to determine the best height for 4x5 inche film above the platen of a scanner? I placed a razor blade on the platen elevated by a piece of cardboard. The green channel looks sharper. However, I couldn't tell much difference along the edge from the platen to the elevated portion.
I checked with the eyedropper. I thought that there would be more light falloff as the out of focus image was diffused, but even so, it only extended a pixel or two as the edge was elevated off the platen. Thank you and best regards.

Mike

[Ted Harris]

Hello! What is the best method to determine the best height for 4x5 inche film above the platen of a scanner? I placed a razor blade on the platen elevated by a piece of cardboard. The green channel looks sharper. However, I couldn't tell much difference along the edge from the platen to the elevated portion.
I checked with the eyedropper. I thought that there would be more light falloff as the out of focus image was diffused, but even so, it only extended a pixel or two as the edge was elevated off the platen. Thank you and best regards.

Mike

Mike, it is partially an exercise in futility. First, it will vary from scanner model to scanner model. One can assume that whatever the height of the film, in the holder isthe height you should use but you need to masure this with a depth micrometer for it to be truly useful since we are talking small measurements. Second, the first point assumes a evel of quality control in these consumer scanners that I am sure doesn't exist. I have never checke on the tolerence levels for placement of optics, mirrors, holders, etc. but it is an interestin gpoint and I will. I am willing to bet that, for either Microtek or Epson, the tolerence levels are wider than any adjustments we might make, meaning that what you are doing could work miracles and could be meaningless.

I go back to what I said in another thread a day or so ago. These machines are marvels of performance for the price we pay for them but it is absolutely ridiculous to expect them to come lose to the performance of machines costing 25 to 50 times as much ... remember tbat 50 times as much. There are similar comparisons all over the imaging world. The cost of consumer video recording versus broadcast recording standards; the cost of a lens for a 'prosumer' digital video camera versus the cost of a set of lens to shoot a major motion picture, etc. In most instances those of us at the consumer end of the spectrum would not only be surprised at the prices of the hardware at the other end, we would be flabergasted.

[Michael Heald]

Hello! I agree that one shouldn't try and make a $500 dollar desktop achieve the resuls of a $10,000 scanner.
Folks have noted in past threads that there are ways to optimize the performance of an individual desktop scanner. Some have recommended finding whether the scanner focuses the best in red, green, or blue for B&W. Others have found a significant variation of performance of a scanner based on film height above the glass. This may mean that there are variations between units of any particular model and that there are no clear recommendations for any particular model of desktop scanner.
However, can these optimizations yield improvement such that there is a noticeable difference in quality of scans for a particular or individual desktop scanner? Best regards.

Mike

[Doug Fisher]

Here are some sites that show examples of the height being varied. Norm's site has a nice discussion of the issue:

http://www.normankoren.com/Tutorials/Epson_flatbeds.html#Focus_spacing

http://www.larry-bolch.com/ephemeral/4870-shims.htm

Doug
---
www.BetterScanning.com

[Michael Heald]

Hello! Thank you for the info. Perhaps a scene with lettering would work better than an inclined razor blade. Best regards.

Mike

[Leonard Evens]

As Doug's links show, resolution of fine detail does depend on heght but the effect is fairly subtle. It also probably varies from unit to unit so you have to do tests for your scanner. Also, a sharp edge from a razor blade is pretty easy to resolve well, so you are not likely to see much difference with such a source. You want something with a lot of fine structure. Also, remember that you are talking about resolution in the range 20-40 lp/mm which is way above what the human eye can resolve without magnification. So even very small print is probably not fine enough. In addition, keep in mind that there are going to be differences between reflective scanning and transmission scanning. Ideally you would want to use a film source with fine detail, perhaps a resolution test chart, way above 40 lp/mm. Finally, the optical and other limitations of the scanner will blur the image somewhat at any height, and that will make it harder.