Continuous Inkjet

Rayleigh worked out in 1878 that a liquid jet cannot stay a jet. Surface tension amplifies any disturbance along it until the column pinches into drops, and the wobble that grows fastest has a wavelength of about four and a half times the jet's diameter. A fifty-micrometre nozzle therefore breaks up into drops at tens of kilohertz whether you want it to or not. Rune Elmqvist at Siemens used that in 1951 to make a chart recorder that wrote with ink instead of a pen. Richard Sweet at Stanford turned it into printing in 1964 by adding the two things that make it controllable: drive the jet with a piezo AT its natural break-up frequency, so every drop is identical and arrives on schedule, then put a charge on each drop at the instant it separates and deflect it electrostatically on its way to the paper. The consequence is a machine that is always firing. Drops you do not want are simply left uncharged, fly straight on into a gutter, and are pumped back to the reservoir. That sounds wasteful and is the single reason the technology survives: a nozzle with ink permanently flowing through it cannot dry out, so a continuous inkjet can sit in a factory for years and start instantly. It is why every date code on every can and bottle you have ever bought was printed this way, at conveyor speed, onto a curved and moving surface. It is also why nobody put one on a desk: the gutter, the pump, the filter, the viscosity controller and the evaporating solvent are the price of never clogging.