[GreenKeys] Teletype (telephone) Local Loops

[Jim Haynes]

Yeah, we have it simple in our ham shacks, but it could be simpler yet.

Telegraph test boards were pretty awful in the early days because each
jack went to a wire on a pole that went somewhere.  If the wire broke
or fell down the board man had to find a spare wire to replace it, and
that spare wire might not have anywhere near the same resistance as
the one he was patching around.  So loop currents had to be adjusted.

There is also the real vs. virtual aspect of telegraph circuits.  In
old railroad days the loop simply passed through all the stations along
the line, so any one of them could send to all.  But with business use
of telegraphy the lines went from telegraph/telephone office to office,
with side branches to the customer sites.  So it became the practice to
use repeaters in the offices to feed the customer sites, the repeaters
creating a virtual loop through all when the actual circuit might be wire
part way and carrier part way and didn't go anywhere near the customer's
premises.  And you might want to use polar on the main line for signal
quality reasons, but it takes some trickery for a polar circuit to be
operable from any station along the route.

Western Union developed a way to simplify things they called PCH for
Polar Centralized Circuit Handling.  See Western Union Technical Review
8:1, January 1954, 8:3, July 1954, and 12:1, January 1958.  Incidentally
Western Union Technical Review in my mind is about equal in importance
to the AT&T "green book".  In a PCH office every circuit has a send leg
and a receive leg, and things are standardized so that any pair of legs
can be connected to any other pair without having to adjust currents.

I'm less familiar with the Bell System approach to the same problem, but
it involved a concept of "hubbing" so that circuits could be patched
together without consideration of loop current.  Both the Bell and W.U.
schemes included repeaters that could be used to create virtual loop-
through multistation circuits.

Then when we got transistors things were developed operating at lower
voltages, typically plus and minus 6 volts as in MIL-STD-188 and
EIA RS-232.  The 60ma loops where they still existed were quite
separate from the test boards that ran the low voltage stuff.

For a while I operated my station on a voltage basis rather than a
current loop basis.  This involved vacuum tubes as selector magnet
drivers, so I had 0 volts for mark and -40 volts for space.  An
inconvenience of this was that I couldn't just plug in a plain TTY
machine into the system without having a selector magnet driver and
keyboard signal generator set aside for the purpose.

Still later, when we got good high voltage transistors, I started on
a similar system using +/- 6 volts.  But I am just too lazy to have
fully carried it out, so I still had a mixture of current loops and
voltage signals.  And every manufactured TU that came along was
built for current loop and had to have an adapter or be reworked to
fit into the voltage scheme.  But it sure simplifies switching - no
need to break into a loop to add a printer, nor to keep every printer
in a loop so it doesn't run open, nor to allow two different senders
to send to a hub provided they don't both do it at the same time.

Jim W6JVE

 	---

 	"Ya can argue all ya wanna, but it's dif'rent than it was."
 	"No it ain't! No it ain't!  But ya gotta know the territory."
 		Meredith Willson, The Music Man