[MacDV] Re: Film Scanners (slide feeders)

[Danny Grizzle]

On 12/31/02 12:02 PM, "Randy Wilson" <WilsonR at fonix.com> wrote:

> It's too bad the slide feeder for the Nikon seems to have so much
> trouble.  I have a 1957 "Kodak 300" slide projector that has a great
> feeder.  You slap 80 or so slides into the front side, and you can rifle
> through slides every second or two if needed.  I have run about 10,000
> slides through it in the last couple of months and have had only about 5
> slides get stuck in it.  Maybe Nikon should look at some 1957 technology
> and see how it was done?

The difference in a slide scanner and a slide projector is in precision.

In common carousel type slide projectors, slides are gravity fed into
position. They drop into place, and are then pinched by two metal fingers on
the edge of the slide mount to secure position of the film plane, which
controls focus. On pro Ektagraphic models, slides are also pinched into
position laterally on the horizontal edges, to repeatedly position the slide
mount aperture so that the audience does not experience border shift during
dissolve-type transitions between slides in two projectors.

The emphasis is on fast image changes. I grew up in Longview, Texas, where
Eastman Kodak operates a chemical plant. Kodak once produced an annual
multi-screen, multi-media extravaganza, a monster AV tour to highlight Kodak
products and top tier, National Geographic-level photography. Thanks to
their involvement with my hometown community, I was able to attend this show
several times as a teenager. I always thought this show mostly featuring
still images was as effective as a major motion picture. It certainly moved;
sometimes the pace of the show sparkled with slide changes occurring rapidly
like lighting a string of firecrackers.

Film scanners, on the other hand, have a moving slide stage that is
positioned in microscopic increments. In Nikon 35mm scanners, at least, the
CCD sensors are stationery, while the film stage is moved by a very high
precision stepper motor during the scan. Consequently, slide loading is very
slow by comparison to a slide projector.

The innards of a Nikon auto feeder have similarities to an old fashioned
slide projector advance mechanism. Slides are loaded in stacks of about 40
slides, all oriented horizontally regardless of picture composition. There
are two slide trays, one the supply, the other to catch completed slides.
Instead of a push slide (the reason we call the "slide" projectors), the
Nikon auto feeder uses rubber rollers which grip the wide sides of the mount
to feed slides into the mouth of the scanner. Rubber rollers also take hold
of the slides that are ejected from the scanner, pushing them into the out
tray.

Actually, in addition to the rubber rollers, there is also a plastic "push"
shim, which drives slides into the scanner. This is propelled, I believe, by
a flexible nylon strip gear, very similar to the mechanism in most
automobile power windows.

The auto feeder unload/load cycle is really slow, requiring almost as much
time as the scan itself.

The whole device is very jam prone. It worked fairly well when new, but soon
stopped being reliable, even though subjected to light use. There are some
light metal finger parts that work like the epiglottis in your throat.
Slides in the scanner itself are bi-directional, going in and out the same
opening. The springed metal finger mechanism, about the thickness and
tension of a light gauge guitar pick, sorts ejected slides into the catch
bin, but easily bend to allow slides from the supply bin to brush past on
their way into the scanner.

Jams occur due to both the rubber rollers and the metal sorting fingers.
Typically, what happens is a slide being ejected does not quite clear the
metal fingers, failing to fully seat in the eject bin. Then, as the auto
feeder transitions into loading the next slide, the new slide catches the
trailing edge of the previous slide, dragging it along back into the
scanner. Usually, the scanner completes a worthless double-slide scan, then
jams when it tries to eject both slides sandwiched together.

Of course, this constant interruption disrupts any effort to scan slides in
order, or to utilize consecutive file naming capabilities of the scanner
software. So it creates tons of organization problems, rescanning, etc.

This happens with all types of slide mounts, cardboard or plastic. Rubber
feed rollers do not grip cardboard slide mounts great at all times, and poor
mounts (separating glue, dog-eared corners) are a problem. Plastic mounts
tend to be thinner, and thus more prone to not being fully ejected, multiple
slide jams occurring in scanner.

I've tried everything, such as cleaning rubber rollers (brief relief) to
springing the metal finger separator slightly. Nothing seems to work.

Bottom line: the device can't be trusted to do work unattended. Since it has
to be supervised full time, I figure the device was a waste of nearly $600
because the task could be accomplished by an operator manually nearly just
as fast.

Danny Grizzle