A trivial point of history

[From Bill Powers (2000.09.08.1459 MDT)]

Martin Taylor 2000.09.08 10:58--

When we talk about the origins of the negative feedback idea, we
often start with H.S. Black. I'm currently reading a book: "The
inventor of stereo/the life and works of Allan Dower Blumlein, by
Robert Charles Alexander, Focal Press 1999. On p140, I find this
paragraph:

------------
On 18 September 1933, Blumlein and Henry Clark applied for Patent No
425553 which specified by name the use of a negative feedback loop in
an amplifier, to reduce the output impedance of pentode valves. This
circuit is especially useful as it improves upon a theory which had
been known since January 1932, when Nyquist of Bell Labs had outlined
the general conditions under which an amplifier becomes unstable when
feedback is applied. Negative feedback is generally attributed to
H.S.Black who outlined the method in January 1934, though the
Blumlein and Clark patent however, predates Black by some four months
with a method that combines both current and voltage feedback. Black
had used an identical formula to the one which Blumlein and Clark
came up with, but had only calculated the linearity of the voltage
gain.

H. S. Black actually figured out the negative feedback equations in 1927,
on the way to work at Bell Labs on the Lackawanna Ferry. He scribbled the
basic equations in blank spaces in his New York Times. For some reason that
I have never heard, he delayed publication for 7 years! I have seen
photocopies of the very newspaper in question, but I'm afraid I have
forgotten where or when.

Best,

Bill P.

[Martin Taylor 2000.09.08 10:58]

When we talk about the origins of the negative feedback idea, we
often start with H.S. Black. I'm currently reading a book: "The
inventor of stereo/the life and works of Allan Dower Blumlein, by
Robert Charles Alexander, Focal Press 1999. On p140, I find this
paragraph:

ยทยทยท

------------
On 18 September 1933, Blumlein and Henry Clark applied for Patent No
425553 which specified by name the use of a negative feedback loop in
an amplifier, to reduce the output impedance of pentode valves. This
circuit is especially useful as it improves upon a theory which had
been known since January 1932, when Nyquist of Bell Labs had outlined
the general conditions under which an amplifier becomes unstable when
feedback is applied. Negative feedback is generally attributed to
H.S.Black who outlined the method in January 1934, though the
Blumlein and Clark patent however, predates Black by some four months
with a method that combines both current and voltage feedback. Black
had used an identical formula to the one which Blumlein and Clark
came up with, but had only calculated the linearity of the voltage
gain.
------------

It's interesting to note that Blumlein and Clark's objective was
impedance reduction rather than control. When I first encountered
what became CSGnet, some of the discussion centred on the issue of
impedance, and I wonder why that aspect of feedback has been absent
from more recent discussions. It's another way of looking at the
(in)ability of external disturbances to influence the value of an
environmental variable. Different ways of looking at things sometimes
lead to novel insights.

Martin