haynes@ucscc.UCSC.EDU (Jim Haynes) (03/28/89)
A certain former manufacturer of computers (which shall remain unnamed,
but is renowned as a producer of light bulbs and steam turbines) made a
lot of tape drives of a novel type. Instead of the familiar vacuum columns
with air pressure switches or photoelectric means to sense the tape loop,
they used "scramble bins" about 8 inches square and as thick as the tape
is wide. The tape goes into and out of the bin, and as long as the bin
doesn't get too full the tape doesn't get creased or tangled; it just
forms a bunch of S-shaped loops in there. To measure the fullness of
the bins they used lights behind them and solar cells around them.
As the bins fill they get progressively darker, and this turns out to
be proportional enough that the reel servos actually work.
In fact, mechanically, the things worked just fine; hardly ever did
the tape come out of the bins entirely or pack the bins too full.
The problem was that tape is magnetic, and with all those loops the
magnetized surfaces were constantly rubbing across one another. Now
when you rub something magnetized against something else magnetized
a likely result is that both get somewhat de-magnetized. So the tapes
could be written okay, and usually read okay the first few times, but
the more times they passed through the scramble bins the less magnetism
they retained, until they became unreadable.
The company ultimately had to design new tape drives with regular
vacuum columns and replace all the units in the field.
haynes@ucscc.ucsc.edu
haynes@ucscc.bitnet
..ucbvax!ucscc!haynes
"Any clod can have the facts, but having opinions is an Art."
Charles McCabe, San Francisco Chronicle