[GreenKeys] RS232 to 60ma loop
Jim Haynes
jhhaynes at earthlink.net
Wed May 10 18:03:47 EDT 2023
Depends on what historical properties we are trying to preserve, of
course.
One of my first assignments as a summer student at Teletype Corp. was
to work on a solid state replacement for the polar relay. What came out
of that was the low-voltage selector used in later model machines. My
thought was that high voltage transistors are rare and not very good
and power transistors are common and good, so why don't we just rewind the
selector magnets for a 12 volt 600 ma loop keyed by a power transistor.
Someone else improved the circuit by taking advantage of the constant-
current properties of the transistor rather than having a simple R-L
saturated transistor circuit. It wasn't original to me; my boss had
thought of the same thing (but didn't mention it to me at the time).
But my big mistake was failing to ask just why we wanted to replace the
polar relay. There were then and later solid state relays on the market,
but I recall from MIL-STD-188 that a real polar relay has some properties
that are hard to duplicate electronically. The MIL-STD calls for the
relay replacment to be "side stable" meaning with little or no current
through the windings the armature will remain where it was left, marking
or spacing. Which requires that the solid state circuit include a flip
flop to hold the state. Also at the time we didn't have optoisolators
to provide the isolation between relay coils and contacts that a real
polar relay has. If I had asked the right question I might have decided
that we should keep the polar relay but eliminate whatever were seen as
its shortcomings.
My other failing was not trying hard enough to sell the concept that I
was working on, that the interior of the set should work on low voltage
signals, and no series loops. Let every selector magnet have a solid
state driver, and every keyboard or signal generator be arranged to
generate a low voltage signal something like + or - 6 volts. RS-232
had not been done yet, but the concept of a low voltage polar signal
was attractive. Would have made it easy, for example, to sell a solid
state regenerative repeater working on transistor voltages without
having to have inputs and outputs as high voltage loops. I was told
that probably no customer was interested in a regenerative repeater
because it would be so costly. I felt otherwise because of my RTTY
experience and the problems caused by missing stop pulses and false
start pulses with radio signals. I expect the military would have
bought lots of them.
The selector magnet drivers would be designed so that with the input
disconnected they would go marking, thus preventing anything running
open. Consider how much simpler it would be to handle a reperforator.
With current loops you have to keep the reperforator in a loop when it
is not in use so it doesn't run open and punch blank tape. To make it
copy you have to insert it into a loop, which means opening the loop,
disconnect the reperf from the loop that keeps it from running open and
insert it into the traffic loop you opened. With the low voltage set
design you just connect the reperforator input to the line carrying
the signal and it starts copying. No need for normally closed jack
contacts or relays with make-before-break contacts and all that.
And it remains a mystery to me why Bell Labs did the 101 Data Sets
with current loop, albeit low voltage, interfaces to the selector
magnet drivers in the Teletype machines.
I built several sets for my own ham station the way I thought they
should be, but eventually quit doing it because it was so much work
to build the selector magnet drivers and I couldn't just take an
ordinary Teletype machine and plug it into the system without having
a current loop driver for it. And there were some historical reasons
for keeping the old loop scheme.
Actually the first version of this I built used tubes rather than
transistors, so the voltages inside the set were grid voltages
for the selector magnet driver tubes. At the time there were still
no good high voltage transistors for driving 120V selector magnets.
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
More information about the GreenKeys
mailing list