>

Jim Haynes  tell us: Well, not really, because the principle of operation of the Inktronic is
completely different from that of all the later ink jet printers.

It was invented by Chuck Winston, who had the idea of electrostatically
deflecting droplets of ink to draw characters on paper. The first patent
is 3,060,429 filed for in May 1958 and issued in October 1962. This
shows the principle using one nozzle writing on a moving paper tape. I
saw that demonstrated in the summer of 1958. If there had been a market
for a high speed tape printer it might have been a real winner.

By the early 1960s there was a demonstration model of a page printer
which used a magnetic core memory to store the bit patterns. The
demonstration model was housed in a big wooden cube and had the ability
to print Japanese as well as Roman. I'm not sure where this was shown,
but places like the Armed Forces Communications-Electronics Assn.
conventions are likely. This used 40 nozzles to print 80 columns on
8 1/2" wide paper.

In the same time AT&T had a contract for what was called Long Lines
Project 176. This was a communication system for a government agency
that involved 100 WPM TTY, 2400 WPM TTY and facsimile, all encrypted
and the stations installed in secret locations. The high speed ASR
set provided for transmission and reception on paper tape and printing
using Inktronic printers. Magnetic core memories were used to hold
the character patterns for the printers. The sets were extensively
RF shielded and also shielded for sound emanations. Deliveries started
perhaps in late 1966. The printers caused considerable hair-tearing -
they pretty much had to be hand-tweaked to get acceptable printing.
One problem was that droplets came out of the nozzles in different
sizes, so that the deflection sensitivity varied from one droplet to
the next. The ink had to be heated to a certain temperature to control
its viscosity.

In the same time frame Hewlett-Packard produced a strip chart recorder
using electrostatic deflection of ink; but only in one dimension. I
believe H-P used ultrasonic pulsing on the ink supply to help produce
more uniform droplet size.

The commercial Inktronic was under development as the 176 project was
winding up. This was to be limited to 1200 wpm and used a magnetic
core read-only memory to hold the character patterns. The main product
was a receive-only printer, but a very few KSR sets were produced.
These used the same keyboard as the Model 37/38, another disgrace to
have such a crummy keyboard in such an expensive machine.

These printers used 40 nozzles to print 80 columns. The same deflection
voltages were applied to all 40 positions in parallel, with a valving
electrode selecting which nozzle was to print at any time. On the whole
they worked well when first put into service and went downhill from there.
Paper dust would accumulate on the electrodes - naturally, since there
were high voltages to attract the particles. Nozzles would get clogged.
It was an exacting job to clean nozzles and deflection electrodes without
damaging them. Maybe someone who worked on these machines in Bell System
service can tell us more about the maintenance issue.