Interesting to read this from the other side of the pond. At the same time here in Germany we stared fascinated towards USA with brands like Klipsch, Electrovoice, JBL, RCA, Motorola, National Semiconductor, Texas Instruments, Raytheon... - certainly far above any Grundig, Saba, Metz, Telefunken, Graetz, Siemens, ITT, Valvo or RFT etc.🙂
Wasn't that more like very late 70s and rest of the 80s?
With all the fancy new horn and PA systems etc.
Guys like John Eargle were also with JBL.
Electrovoice is still one of the best sounding brand names imo.
Yeah, we're talking about a different era. By the late 1970s the use of global feedback in audio circuits was virtually universal, even in "cheap" products.
That answers my question about the patent for amplifier feedback.
That must be one of the earliest applications in a consumer audio product. It's quite convoluted too; there's an inductor in one of the loops. That seems like a bad idea in a high impedance circuit.
The output transformer was a big disincentive to using global feedback. There's two poles to deal with and motorboating used to be a thing with audio amplifiers. Large amounts of negative feedback became much easier to employ with transistorized circuits.
In the 1970s I built an ultalinear amplifier with salvage parts. It was heavily based on a Knight schematic. By using a transistorized power supply, and a much more reasonable 100K input impedance, I managed to make an amplifier that was as quiet as a solid state amplifier (NO hiss, clicks, thumps or hum) without the awful grittiness that most transistorized amplifiers offered at the time.
And it was then that I realized just how important global feedback was in obtaining consistent, repeatable performance. I remember when a balance control was mandatory. Why? Because as tubes wore out, one channel would get fainter than the other until it almost faded away. With global feedback, performance remained consistent until most of the useful life of the tube was already used.
That must be one of the earliest applications in a consumer audio product. It's quite convoluted too; there's an inductor in one of the loops. That seems like a bad idea in a high impedance circuit.
The output transformer was a big disincentive to using global feedback. There's two poles to deal with and motorboating used to be a thing with audio amplifiers. Large amounts of negative feedback became much easier to employ with transistorized circuits.
In the 1970s I built an ultalinear amplifier with salvage parts. It was heavily based on a Knight schematic. By using a transistorized power supply, and a much more reasonable 100K input impedance, I managed to make an amplifier that was as quiet as a solid state amplifier (NO hiss, clicks, thumps or hum) without the awful grittiness that most transistorized amplifiers offered at the time.
And it was then that I realized just how important global feedback was in obtaining consistent, repeatable performance. I remember when a balance control was mandatory. Why? Because as tubes wore out, one channel would get fainter than the other until it almost faded away. With global feedback, performance remained consistent until most of the useful life of the tube was already used.
No one claimed global feedback did not exist in the 50’s and 60’s. It was discovered in the 1920’s after all (Harold Black, Bell Labs). What was said is that it was misapplied when the transition was made to solid state amplifiers and some very real examples demonstrating that were cited but I’ll repeat them here:-
1. Distortion wars amps of the 1970’s (TIM, SID).
2. The whole Matti Otala thing. If ever you needed an example of getting feedback wrt audio amplifiers wrong, it’s his papers on the subject. See Dr. Putzeys ‘F-Word’ refutation of the whole thing
3. In 1998 (and others even later than this) Martin Colloms (BSc Engineering qualification I am told) claimed feedback goes ‘round and round’
4. You only have to root around on Stereophile (think of it what you may) to read interviews with designers claiming the deleterious effects of feedback
Let’s not go down the ‘Waly road’ where things are purposefully misconstrued in order to promote argument for argument sake.
🙂
Here is a link to Black's original article about feedback published in the in-house Bell Labs magazine
https://hifisonix.com/wp-content/uploads/2010/10/Bell-Feedback.pdf
https://hifisonix.com/wp-content/uploads/2010/10/Bell-Feedback.pdf
Loop probes can be useful with DACs too. For radiated RFI, usually more than for oscillations.
Designers that don't understand feedback will in fact experience deleterious effects in their designs.
Thank you Jan, looking forward to more articles. Some of it will be over my head, but so what, that is how you learn. Nobody can be an expert in everything.
I used to use them in high voltage and short circuit testing labs to measure fast dH/dt fields. Shielded electrostatically.
LTP Magic
Lets assume the greatest error results from the transfer function at the base or gate, then a one die LTP where the amp is used in non inverting mode would generate the least error
Lets assume the greatest error results from the transfer function at the base or gate, then a one die LTP where the amp is used in non inverting mode would generate the least error
Totally agree, although to some extend it's also trying to understand each other.
Which can be quite the challenge wading through layers of being just text only, background differences, cultural difference, language differences, age differences and personalities.
To end on a positive note, even though it can get pretty hairy and rocky here, I am actually amazed how much faster things go than most companies I've worked for.
Which is interesting, because most of them don't have the issues of text only and often there isn't much of a big difference in culture between people.
My table radio has its 35L6 output tube set up with just an unbypassed 150-ohm 1 watt carbon resistor to gnd.
Just as an aside, there is much discussion of when feedback really began. My take is that the dawn of feedback never really came into its own until man began walking upright... before that they just kept falling over...but for the Irish it took a little longer