This may seem a non-sequitur to some. After all, I don’t post designs on this website that make use of feedback. This is simply because I don’t like the way feedback affects the sound of an audio amplifier. But that doesn’t mean that I don’t understand feedback and make use of it in other applications.

There is no easy way to say this, so I’m just going to accept that I’m about to offend a bunch of people. Most of the discussion about feedback and feedback amplifiers in audio forums on the internet is not only incorrect, the vast majority of the advice and discussions demonstrate a profound ignorance and misunderstanding of feedback amplifiers, how they operate, and what “stability” actually means. This also applies to some fairly widely read books currently published on vacuum tube amplifier design.

Employing feedback does not mean just adding a wire from the output to the input. It is not a simple fix to make something stable. And it is most definitely NOT something which should ever be switchable. A feedback amplifier is something that is designed, from the start, to employ feedback. The proper design has impacts on not just the feedback elements, but the design of the forward amplification path as well. The entire “system” that is the amplifier is designed differently in feedback verses non-feedback amplifiers.

The reason we have feedback amplifiers is because of a simple paper written in the 1930s by a Bell Telephone Laboratory, Inc. employee, who was also a member of the American Institute of Electrical Engineers or AIEE (a forerunner to the current Institute of Electrical and Electronics Engineers or IEEE), named Harold Stephen Black. In January of 1934 his most famous paper was published in the Journal of the A.I.E.E. under the diminutive heading “Stabilized Feed-Back Amplifiers”. (Full citation: H.S. Black, “Stabilized feed-back amplifiers”, Electrical Engineering, vol. 53, pp. 114-120, Jan. 1934).

I have taken the trouble to post the full paper here so that anyone can read it. And I encourage anyone working with (or simply curious about) feedback amplifiers to do so. The paper contains only simple algebra, a few complex numbers, and a little bit of simple trigonometry. It should be fully understandable by anyone with a high school level mathematics education.

I do this because this paper is the original thought from which all feedback amplifier design springs. I want people to read and understand this paper and its design concepts. Without such understanding, feedback amplifier design and operation will remain a mystery. If more people would simply read this paper and make an honest effort to understand it, there would be far less incoherent babbling concerning amplifier feedback across the internet. It’s clear that many of those discussing feedback (including some fairly famous people) have either never read this paper, or read it and completely failed to understand it.

Amplifier feedback is not magic, nor is it mysterious. However, it is also not an ‘add on’ or a ‘fix’. It is also topology independent and all works the same way. And it’s employed in amplifier design to linearize otherwise non-linear response functions. That’s it. However, I will say this: If your goal is simply a very linear amplifier, there are far easier and better ways to achieve that goal than using vacuum tubes.

Note: The design and selection of criteria by which to design feedback amplifiers is beyond the scope of this simple post. I can go into such topics if anyone is interested however, many should become self explanatory with a thorough reading of Black’s 1934 paper.

As always, questions and comments are welcome.