[GreenKeys] [External] Tip on loops

[Harold Hallikainen]

> On Oct 4, 2019, at 2:28 PM, Jim Haynes wrote:
>
>> ... The reason for this is the temperature
>> coefficient of resistance in the metal filament.  As current increases
>> the filament gets hotter, increasing its resistance and tending to
>> reduce the current.
>
> I vaguely recall some interesting circuits that I recall seeing in the
> 1960s that used incandescent bulbs as active elements, exploiting both the
> non-constant resistance of the bulb and the thermal time-constant of the
> filament.
>
> 		Doug Jones

This was used in early HP audio generators. An incandescent lamp adjusted
the gain in a twin T oscillator. At the output went up, the lamp got
brighter, the resistance went up, and the gain went down. This allowed the
oscillator to operate at just the right loop gain instead of clipping if
the gain went too high or dying if the gain went too low.

I think these were also used in AM receivers in the 1930s as a volume
expander for noise reduction. Putting a lamp across the audio resulted in
loud audio increasing the brightness and increasing resistance,
attenuating the audio less, and making loud sounds louder. Quiet audio did
not light the lamp brightly, so the resistance went down increasing the
attenuation, making soft sounds softer.

Tubes designed for this purpose (current regulators) were "ballast tubes"
as described at
https://www.americanradiohistory.com/Archive-Bookshelf/Author-Groups/Jacobi/Jacobi-Ballast-Tube-Handbook-opt.pdf
.

Use of a ballast tube in a Teletype loop would be interesting. I think it
would adjust the loop current to average at the rated current (such as 60
mA) while we want the current during mark to be 60 mA and during space to
be 0 mA. If a 60 mA ballast tube were used, I think the mark current would
be high since the tube is averaging over both mark and space.

Harold

-- 
FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.