[GreenKeys] Loop Supply

[Harold Hallikainen]

On Mon, January 20, 2025 8:46 pm, Jim Haynes wrote:
>
> See also
>
> QST Issue: Jan 1984
> Title: Universal RTTY Current-Loop Interface, A\

This circuit does not use any active current source or anything fancy.
Instead, it just supplies a way to key the loop and a method of sensing
the loop. This is similar to the TU-300 loop interface at
https://w6iwi.org/rtty/tu300/LoopDriver/Tu300LoopDriverSchematic.pdf
except that it is polarity sensitive

>
> QST Issue: Jan 1972
> Title: Modern Teleprinter Local Loop, The

Looking at this article, it appears there is a maximum of 24V across the
selector magnet. This will not make the current ramp up as fast as 100V.
With the high voltage and big resistor of "traditional" loop supplies. It
would be interesting to plot the selector magnet current with a
traditional loop supply and with this active loop supply. The active
supply does maintain the current as the magnet voltage increases, but so
does the traditional supply since the magnet voltage is a small part of
the applied voltage.


Selector magnet inductance, voltage across selector magnets, etc. possibly
brings up another subject. I've been asked to design a Keyboard Operated
Send for the TU-300/TU-170A. In the DSP TU (
https://w6iwi.org/rtty/DspTU2/ ), I sense loop current interruptions to
trigger KOS. However, the loop current is ALSO interrupted by received
data. In the DSP TU, I lock out the KOS when a space is sent to the keying
SSR by the received data.

However, I do not have ready access to such signals from "outside the
box." My current thinking (subject to test) is to not sense loop current
for KOS, but, instead sense the voltage ACROSS a machine (keyboard and
magnets in series). When the keyboard is used, the voltage across the
machine will go up to 50 volts or higher, depending on the loop voltage.
That voltage will stay high for at least one bit time. If the loop is
keyed by something else, such as a terminal unit, there will be a flyback
high voltage from the selector magnets, but that should disappear well
before a bit time. So, I'm thinking of making a voltage divider with large
resistors to avoid placing a significant current path around the selector
magnets, and have that drive an opto like the H11AA1. I expect to see long
pulses (at least, say, 11 ms long) when the sensed keyboard is activated
and very short pulses when something else keys the loop. These would be
distinguished by a cheap microcontroller to drive PTT and the TU transmit
line with a user configurable timeout.

Another project!

Harold
https://w6iwi.org



-- 
Not sent from an iPhone.