Inexpensive film/slide scanner needed. Suggestions ??

Discussion in 'Scanners' started by Ian Woodrow, Mar 6, 2006.

  1. Ian Woodrow

    Ian Woodrow Guest


    Here's the story...

    I acquired a used Nikon Coolscan III (LS-30) but the results weren't so
    great. The initial images are very soft in a way that would indicate
    possible dust on the inside mirror. So, knowing that to get this
    serviced would cost a fortune I had a look around and found a guide to
    dismantling to reach the
    mirror. Well, there were some screws that were just STUBBORN and would not
    budge. The guide
    I saw was for the LS-2000 but on trawling the good ole web I found a report
    from someone who
    has the LS-30 and says you can reach this mirror without totally
    dismantling. Well, might as well give
    it a go. NO joy. Results are still soft so it's not looking good.

    I am now toying with the option of having to buy a slide/neg scanner without
    paying an absolute mint.
    I'm looking for something that will give me sharp results and be good for
    viewing on a PC monitor or
    printing maybe about A4 size. Including the possibility to crop and still
    get good result.

    I'm sure there are probably some out there - not of which will be ADVERTISED
    as 'not great', but
    ARE in fact 'not great'.

    Can anyone point me in the right direction for something that does the job
    well but is not in the ... 'very
    expensive for something you don't really need' range. I don't really need a
    slug's eyeball crop to be
    pin sharp at A3.... good results cropping to half a slide frame would be
    nice if sharp on screen.

    ANY hints at all welcomed - APART from buy a digital SLR !!!


    Woody, around Glasgow in Scotland
    Ian Woodrow, Mar 6, 2006
    1. Advertisements

  2. Ian Woodrow

    Dave Guest

    There's only one option in my opinion. That would be the Minolta Scan
    Dual IV. They are available in the US from NewEgg for $220 which is
    about as low as you can find for a decent scanner.

    This is not a workhorse type of scanner, but it is the one I use for the
    few slides and negatives I need it for. 3200 PPI is plenty for at least
    8x10 IMHO.

    Hope this helps,
    Dave, Mar 7, 2006
    1. Advertisements

  3. Ian Woodrow

    Ernst Dinkla Guest

    Typical Aztek sample. You need a warped slide and the wrong
    focus spot to achieve a result like that on 35 mm slides.

    I have pioneered the wet mount method on the Coolscan 8000 and
    used all the tweaks to get the best results. Before Grecco and
    Aztek. I can assure you that the quality gain with wet
    mounting is important but the samples shown are more an
    example of bad use of the normal carrier than the advantages
    of wet mounting (if the good scan represents the actual
    information in the film).


    Ernst Dinkla
    ( unvollendet )
    Ernst Dinkla, Mar 7, 2006
  4. Ian Woodrow

    JuneBug Guest

    Very nice tutorial, but not easy (possible?) to implement on a Minolta
    5400 carrier. The glass from a 35mm Gepe mount is too small to put a
    piece of tape on. Likewise and to a less degree, the holder also does
    not have enough room for the tape. What would be nice is to have a piece
    of glass that is exactly the same size as a 35mm slide mount, and can be
    dropped into the slot of the holder. No taping would be required.

    I ran into lots of problem with a Gepe anti-newton mount. Removing
    finger prints and dust from the glass is tedious at best. What is a good
    approach to do this right?
    JuneBug, Mar 7, 2006
  5. Ian Woodrow

    RichK Guest

    Well, you're addressing a question I've asked before on film/slide scanners.
    How do you keep the film flat, unless you place it between two glass plates.

    Slide projectors did some compensation with the projection lens, but I have
    not seen this mentioned anywhere on scanners.

    I do not own a scanner yet and am considering the one Dave (here) mentioned
    Minolta Scan Dual IV. But I'm still bothered how do scanners manage to
    compensate for the curvature of the film/slide in order to keep the results

    RichK, Mar 19, 2006
  6. Ian Woodrow

    RichK Guest


    Wonder why you address Kodachrome specifically? Are not all the slide
    films (basically two) the same. Or is the Kodachrome much worse than the E6

    RichK, Mar 19, 2006
  7. Ian Woodrow

    RichK Guest

    Is it for the same reason that focusing is less successful? Or are you
    addressing only dust detection?
    Would flattening the slide help any with dust detection?

    RichK, Mar 19, 2006
  8. May I jump in here? Focusing on the entire slide is the primary
    challenge of shoe-box scanners, which is why their software either
    focuses on the middle, or *you* choose where to focus.

    This is why drum scanners are superior. They bend the tranny in one
    dimension so when it passes over the CCD or CMOS chip, the distance
    from the chip to the film media is perfect. Check out how Imacon
    scanners work... Termed as "virtual" drum scanners, they flex the
    film by sandwiching it between layers of flexible magnetic pads and
    rotating it over a curve... hence keeping the film in an exact
    distance from the chip as it traverses through the scanning process.

    With mid-range pro-sumer film scanners, I'd go for a Nikon or top-end
    Minolta. Their optical quality partly makes up for film curvature
    focusing issues.

    Otherwise, get your "keepers" scanned at a reputable lab on a drum
    scanner. DPI ratings aren't everything.

    Jim's $0.02.
    jim.hutchison, Mar 23, 2006
  9. Ian Woodrow

    RichK Guest

    I was hoping someone would jump in here :) This question has bothered me,
    from day one, when I heard of film/slide scanners.

    Unmounted filmstrip can ptentially be flattened by the scanner. But not
    knowing the mechanics of any of them, I wonder if any actually do flatten
    the film, and how well. There's plate in a camera to flatten the film, an
    it does a reasonable job of it. Slide projectors had special lenses, but
    that did not work real well either in bad cases.
    This solves majority of the curvature problem, to some extent. The surface
    of the slide is not like a part of a cylinder, but a complex surface. The
    only way to get it 100% flat, would be to remove the film from a mount and
    place it between two glass sheets.
    Seems the only way to compensate would be to measure the distance to the
    film with every snapshot of the scanner, but the sensing device can be as
    wide as the film, so that would not work.
    I agree with you that DPI is not everything, this is why I raised this
    question. It's never addressed in literature or discussions.

    RichK, Mar 23, 2006
  10. Actually, it is addressed: take a look at Imacon's scanner here:

    A picture is worth a 1000 words. Play the movie, and you'll see how
    in fact bending the film does guarantee a consistent distance to the
    lens/CCD assembly.
    jim.hutchison, Mar 23, 2006
  11. The objective of the exercise isn't to get the slide flat but to have
    the surface of the slide maintain a known profile. On a drum scanner,
    which scans a single point at a time on a cylindrical surface, the film
    is deliberately bent in one axis to match the curvature of the drum.
    This action ensures that the other axis is maintained "flat", so that as
    the drum rotates and the spot gradually progresses across the slide, it
    stays at a fixed distance from the optics. If you bend a sheet of
    flimsy paper into a cylindrical section, you can see the same effect as
    the drum scanner achieves - in the other axis from the bend, the paper
    is flat and rigid. So you don't need two glass sheets to achieve good

    However, Jim's comment is only partially correct. Whilst this is
    certainly how a drum scanner works it doesn't mean that it is the only
    way of getting consistent film flatness. If you look through the film
    feed of the SA21 or SA-30 adapters for the Nikon scanners, you will
    notice that you cannot see straight out of the aperture at the back of
    the scanner (if it is an x000 bulk model). The reason for this is that
    the film path is not straight. The film is deliberately bent on either
    side of the frame aperture and, just as with the drum scanner, this has
    the effect of flattening the film curvature in the other axis. Rubber
    rollers also ensure the film is taut across the aperture, keeping it
    fairly flat in that axis as well. It doesn't work perfectly and some
    seriously curved films still present problems, but I have heard of some
    films that won't lie flat on the drum either.

    As Surfer mentioned though, neither of these techniques work with
    mounted film, which relies on the mount to keep it flat, which is
    generally the least successful of all.
    Kennedy McEwen, Mar 23, 2006
  12. Ian Woodrow

    RichK Guest

    To understand this correctly, you must be referring to a line, rather than
    point. Perhaps the data is acquired, one point at a time, but the sensor is
    made like a line, is it not?
    Follow that, yes.
    Agree, but a quick look at a mounted slide, even with a naked eye, will show
    that mounted slides are not flat at all. This is what I was referring to as
    complex surface. So unless you unmount the slide, no scanner will do a good
    job. The drum scanner is of no help, unless the mount is removed.

    RichK, Mar 23, 2006
  13. Not on a proper drum scanner. The slide is placed on a glass or perspex
    drum that spins around and the point scans one line for each revolution
    of the drum, creating a scan line in one axis. The drum is mounted on a
    screw thread so that as it spins the slide gradually moves past the scan
    point to produce the scan in the other axis.
    That is exactly what I said - the whole point of the slide mount is to
    keep the slide flat, but it is not very successful in doing that, much
    less so than either the Nikon adapter or the drum scanner.
    Kennedy McEwen, Mar 24, 2006
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.