Inferences and Comment from the Table of Printers

1. The output column is the output in feet per minute [fpm], or in frames per second [fps] and this refers to the most commonly used speeds with shrunken film
2. The maximum shrinkage possible column is as reported by the organisation and hasn't been checked. It is inevitable that one archive may accept a result that would be unacceptable to another. Many of the continuous rotary printers shown in the table probably exhibit slippage of the image to a greater or lesser extent.
3. The best quality is theoretically achieved only by variable pitch optical printers using precision register pins, claws and sprockets which match the sprocket holes and pitch of the shrunken film and whose image is enlarged on the print material to be the same size as the original frame before it shrunk. The only printers that can completely achieve this are still to be built! The nearest are optical printers with short pitch movements that matches the short pitch of the shrunken film. The Oxberry, Research Products and Neilson-Hordell come close to this but are slow. The Sigma [continuous optical] has a reputation for being awkward to handle and difficult to set up, but only two were made. The Debrie TAI [step optical] fitted with a variable pull down is capable of 25 fps and is undoubtedly the best machine for shrunken film. These last two are the only printers to have been manufactured specifically to handle shrunken nitrate film [1996].
4. Modifications to normal printers include:

1. [a] interchangeable gates fitted with short pitch sprockets and/or short pull downs.
2. [b] removal of register pins to allow a variation in film position in the gate
3. [c] filing of register pins, claws, and/or sprockets to allow a variation in pitch to be accommodated by a variation in film position in the gate
4. [d] reduction of the wrap around on rotary contact printers. [Flat gates are always preferable to curved gates if the pitch is short]

1. Many archive printers throughout the world simply rely on filed down or worn mechanisms to allow shrunken film to be transported and printed.
2. The only technique reported for assisting short pitch film through standard continuous contact printing gates is pre-waxing to allow the film to slip. Theoretically this could result in low sharpness, as is encourages the films to slip against each other in the exposure gate.
3. Most step printers are slower than the best continuous contact printers. Step printers give steadier pictures than continuous contact printers do in general [but this may not be significant in already unsteady archive film].
4. Wet gate prints are unquestionably essential from most archive film to minimise negative scratches, but the Schmitzer, Ott or any gate with rubber or plastic seals touching the film risks catching the edges of old splices and breaking the film. The slow Oxberry just about manages to cope with poor splices. Only aquarium gates in which the entire head is immersed in the wet gate fluid and the film enters and leaves the liquid through its surface can be used safely at speeds over about 5 fps.
5. The fastest printers used for shrunken film are as follows:

1. Debrie Matipo - only now available second hand, step contact, dry, capable of handling the worst material at speeds up to 50 ft/min. Needs some modifications that can be done by a good laboratory engineer [e.g. removal of register pin/s, shortening or filing edges of pin, and shortening and/or filing down claw edges]. The disadvantages are that it is not possible to convert to "Wet gate" and that modern grading techniques [using analysers and punch tape or frame count cueing] are not easily possible without fitting an entirely new lamp house. Similarly, colour grading isn't economic without an additive lamp house. For all these limitations, the Matipo is still the workhorse for many archives and is probably the only European printer to be seen in the USA because of this.
2. Modified rotary contact printers with or without wet gates. The Bell and Howell Model C, the Carter, and the Peterson are all unidirectional printers used by labs and archives with modified gates and/or sprocket drives. Most rely on "worn down" sprockets, which may make them unsuitable for modern work. Some Model C's have replaced sprockets and rotary gates with mechanisms with shorter pitches. This makes them very suitable for a range of film shrinkage that has to be selected prior to the modification but it isn't easy to change from one range to another. The Schmitzer wet gate attachment combined with a short pitch mechanism is probably a very fast and practical solution but great attention is needed to old splices. Peterson have built interchangeable printing units for their unidirectional aquarium gate printer. It is said to be easy to change back to standard or to another range of shrinkage. The Bell and Howell Modular Printer with its aquarium gate has only a single sprocket roller [at the gate itself] and can be used for shrunken film especially if the film is not too seriously shrunken. This model is theoretically the simplest and easiest to change sprocket drive. The transport is very smooth, which subjectively disguises slippage at extreme shrinkages, trading off some sharpness for steadiness.

1. There are many other printers that are capable of quite high speeds with shrunken film but all these are flat bed contact printers of old design. The problem with these [Vinten, early Bell and Howells, Lawley, Acme and Union amongst others] is that they are now museum pieces, their printer lights are dim and the grading and the cueing systems they use are awkward, slow and unfamiliar to modern graders. Several laboratories report that they have printers like these that are used to make copy negatives of shrunken material where grading wasn't needed. In theory, a modern light valve lamp house could be fitted to these old printers but this does not seem to have been done.
2. High speeds can often mean higher tensions and therefore more breaks at joins, so all high-speed printers need additional time spent before printing, checking, reinforcing or remaking old joins.
3. The Sigma printers used at NFTA are the only two ever made and it has taken some time to overcome the problems of prototype technology.
4. Given a slow speed and the right choice of drive, shrinkages of up to 5% can be printed. However, by the time most nitrate has shrunk to this extent the base has already broken down. The vast majority of all archive nitrate film shrinkage is between 0.8% and 1.8%. This is beyond the method of working on a low curvature rotary contact printer [like a Peterson] and yet to use a slow step optical printer like an Oxberry is often uneconomic. Debrie clearly designed their TAI to fit this requirement, but it is expensive.