Epson 3200-Optimum Film to glass distance

There are complaints from users that get very unsharp scans with the

3200 and usually blame the scanner's optics registration, etc..

On my 3200, once you scan the first slide the slide light stays on

as long as the scanner itself remains on, the heat developed and the

lower RH of the film's emulsion causes the emulsion to shrink and

the film to buckle down sometimes bringing the film base closer to

the glass and often causing Newton rings. Measurements with a depth

micrometer indicate that the Epson frame is designed to position the

film's base at 1.000 mm from the glass. (My

actual measurements, taken on the film holder's MF opening put the

film's base at 0.998, 0.992, 0.983, 0.958 and 1.000 mm. distance

from the glass) The emulsion side will thus be further from the

glass by a distance equal to the thickness of the film base. In

case of Velvia its thickness (emulsion + base) is 0.234 mm. Thus,

the Epson frame's construction positions the emulsion side at 1.234

mm. from the glass assuming a perfect 1.000 mm distance from the

plane to the glass where the film base rests on the film holder.

With MF film, which is slightly thinner this distance with be

correspondingly smaller. I would surmise that optimizing the

machine's resolution will require a reasonable amount of film

flatness. This might be helped by turning the scanner off

altogether while the film is in the holder within the scanner to

prevent buckling. Any comments? Has anyone attempted wet mounting

to insure film flatness?

Julio,

Try reading through this article from Norman Koren. He did some testing with different film-to-glass distances using an Epson 2450 (similar) and published his results here:

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

Good luck!

TH

Tom: Thanks for the link. Because it is almost certain that Epson designed the film holders to provide optimum distance from film to glass to optics, holder measurements indicate the optimum distance from film base to glass. The Richard Davis tests in the Norman Koren site were made on an Epson 2450 not a 3200 and as I have no valid reason to assume that what applies to the 2450 also applies to the 3200, please take that into consideration on my comments to the tests in the Norman Koren site. On the 2450 machine tested the optimum distance is given as 3/32" or 2.38mm instead of the 1mm. that would apply if the distances for the 2450 =3200. If that were the case, this could indicate that the unit's optics were off-register. Richard Davis, who performed the tests gives no indication of how measurements were conducted (top of film or film base) and wheather he used a micrometer. Although he does not say he may have place the spacers one on top of the other and that could lead to deviations as the spacers may not achieved intimate contact at least within reasonable tolerances. He gives 3/64" (1.19 mm)as the thickness of the holder (really the distance from the holder's film plane to the glass) but if the 2450 holder is the same as the 3200 that is off by about 20% from the correct number. This to me suggests that the tests have to be taken with a grain of salt.

Mold making is extremely expensive proposition and such things are usually made to very low tolerances set by data provided by the Epson optical engineers. As such I am inclined to believe that the film holders will be right on and that any deviations will come from registration errors in the Epson optics (though I have no proof of that) or deviations in film flatness. From experience, films are never flat specially when the heated as is the case with film scans. My instincts tell me that lack of film flatness could be the weakest link in flat bed scanners. Because of the close focus in such devices, depth of field may be extremely shallow and that can be seen in the tests in the NK site. The above measurements should be useful to people that wish to make their own film holders but more useful would be a method of maintaining absolute film flatness. Any ideas?

Julio, I expertiment the same problem with the 3200. I don't have the solution but I remove the film from the scanner after the preview, I do the correction then I put the film on the scanner, then I remove again as soon as possible. The problem is that when the film is heated don't turn flat anymore, this is a problem if you need to scan again. Sometimes I scan my slide mounted in GEPE anti newton glass mounts.

Regarding the scanner: Have you noted a red/magenta cast on the scan?