Scanning the past – third update

I finished the 150th colour negative film yesterday and the 4650th image.

Although I’ve not finished all the negatives, it seemed time to look back and, as a result, I’ve decided to rescan the first fifteen to twenty films as the quality isn’t as good as the later scans. The main problem was the way the film strip was placed in the holder. I started with the negatives in the conventional way, emulsion-side up, as a scan would give an image the correct way round. However there was a problem. The negative strips all have a slight longitudinal curl. Putting the negatives in the holder emulsion-side up meant the centre of the film was lower in the holder due to the curl. The scanner lamps (the scanner has two, a conventional white lamp for the main scan and an infra-red lamp which is used in the image processing to remove effects of dust) generate quite lot of heat as they are on and off for an hour for set of negative strips. This heat seems to have the effect of softening the film slightly and, as a result the film droops a bit more and pulls slightly at the sides out of the film holder. So quite a few of the earlier scans have a strip of the film edge showing on one or both sides. This also has the result the scanner colour and exposure value software calculation was incorrect due to a white or black side strip or perforations showing and the image quality was degraded.

After those first batches of films, I tried a film with the emulsion side down so the film curved upwards and used a setting in the scanner program to mirror the image automatically when it was processed. That setting was a lot better. Any softening of the film only meant the curve became less pronounced and the film strips all remained well-framed in the holder.

The holder has a height adjustment and I originally tried to compensate for the possible different focus between the lower centre and higher edges of the image by raising the holder. With the film the other way round this was no longer necessary.

I have found some of the strips of film are not cut accurately between the images. The processors must have made the original prints from a continuous film and them cut them to fit in the envelopes/folders, clipping the edge of some images. For important images it should be possible to recover the full image by “stitching” two images together and a little editing. Hopefully there won’t be many.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.