1980s Matsushita 31-2

Early digital scanners used fluorescent tubes to illuminate a moving portion of the document whose image is projected onto a linear CCD array by a set of lenses and mirrors. Because the sensor cannot discriminate light of different wavelengths, color scans thus required three passes, each under a different primary color illumination. Fluorescent tubes were well suited for this application as they can be designed to radiate specific light spectra determined by the nature of their phosphor. Since color scanning requires highly saturated light colors, color filters were still necessary in some cases in order to remove parasitic light outside certain spectral bands. As a result, and to simplify the scanner design, special blue and red lamps were designed with a tinted glass bulb, like the Matsushita model 31-2 shown here, a source of blue light.

Except for the nature of its glass and of electrodes, this lamp has the same design and construction as those of standard 8 W fluorescent tubes. The inner fill consists of a common argon-mercury Penning mixture and the fluorescent coating is a standard cool-white calcium halophosphate material. Its heavily doped cobalt glass bulb is selective enough to let blue light out while blocking out long-wave light radiated by the phosphor and the mercury discharge. The use of a mass-tinted glass material also ensures a stable spectral output characteristic through life, necessary to ensure a consistent color reproduction.

The use of those lamps diminished gradually during the 1990s as a result of the introduction of dielectric-barrier discharge lamps and of color-discriminating CCD arrays with built-in filters. The development and release of high-brightness red, green, and blue LEDs in the second half of the 1990s eventually marked the end of colored fluorescent tubes in digital scanners.