[AT] OT: the inch is somewhat metric

[pvorwerk at newulmtel.net]

> The railroad I trained with was using a paper punched tape “Search
> Machine”. A cable was wired into each of the newer locomotives and a
> cable with mating multipin connector was wired into the “Search
> Machine”. In phase 1 the “Search Machine” checked resistances of
> components through the wires. In phase 2 the leads to the traction motors
> are tied to a resistor bank and the locomotive operated at various loads
> up to full with the power going into the resistor bank. I have seen them
> glowing cherry red in the winter. Phase 2 measures voltages at various
> points.
>
> If a resistance or voltage is out of its range, electricians looked at the
> locomotive electrical schematics and made measurement with meters. They
> might adjust, repair, or replace a component. The idea was to catch a
> problem before it came to a head with the locomotive hundreds of miles
> away from the shop. Moisture in the traction motors was a common problem.
>
> The company, I think, became part of Textron. The locomotive manufacturers
> may have added similar features as an OEM offering.
>
> [Jim Becker] <snip> On the subject of paper tape.  Ask almost any
> programmer how big a program is, he will give you an answer in bytes,
> kilobytes, or these days megabytes.  Ask an NC programmer about the size
> of a program, he will answer in feet, as in how long the paper tape would
> be if you punched it out.  I found that even in the ‘90s, long after
> paper (or mylar) tape had disappeared from the shops, that was still the
> way they answered. <snip>
> Â 
> [Stephen Offiler ] <snip>The very earliest numerical-control (NC) machines
> had no computer, but rather read instructions from a punched paper tape,
> and these were in the research stage in the 1940's and began to appear in
> practice in the 1950's.  It was not until the 1970's that computers and
> digital displays began to appear on the shop floor.   Conversion from
> metric to inch in the displays was not an issue.  Even if it was, the
> conversion could have been accomplished regardless of the definition,
> because even prior to the universal adoption of 1" = 25.4mm exactly, the
> errors were on the order of 25.39993 to 25.40005 which is well beyond the
> precision of the machines in those days.
> Â 
> Â 
> _______________________________________________
> AT mailing list
> AT at lists.antique-tractor.com
> http://lists.antique-tractor.com/listinfo.cgi/at-antique-tractor.com

When I was attending college at St. Olaf one of the Economics professors
liked to describe the first computer he worked on in the 50's at MIT. The
computer would have filled the entire basement of the Econ building (our
Econ building wasn't huge compared to larger schools, but it still made
for a pretty good sized machine.) The computer had a total memory capacity
of 8k. The first program that they wrote to run on it was for some simple
statistical analysis, but when it came time to run if for the first time
they found that their program slightly exceeded the 8k memory capacity.
After scratching their heads for a little bit, they came up with a rather
simple solution: they eliminated the random number generating portion of
their program and instead manually generated the random numbers by
shuffeling punch cards. Still love that story.

Phil Vorwerk
Somewhere in the wilds of Southern Minnesota
>