From a design point of view, not having a output transformer was the biggest win of all I think.
They are either big, expensive and clunky, or you have to compensate so much that you either clip your front stage, or just saturate the transformer.
But yeah, having so much gain sounds nice in theory. I practice it often means that just farting next to it will sometimes be enough to start going in bezerk mode.
My 1950's valve radio has a pentode with decoupled cathode resistor as its output stage, like many pre-1960 radios have. It is anything but HiFi, but I like listening to it anyway.
I agree with this. It seems to be all or nothing in audio - and not just with distortion because people are now even talking about DACs with 27 bits of dynamic range. The correct way is probably somewhere down the middle. I’ve designed and built very low distortion amps and not so low distortion and both types give me as much listening pleasure. If one insists zero distortion is the only way, I’m afraid you will suffer from ‘objectivist fantasy vs subjective reality’ to quote John Atkinson. One of the best sounds I ever heard was a big Pass amp driving a pair of KEF Blades - NP is not pathological about THD.
(Did you manage to get the Leonard Bernstein recording by the way?)
Yes, I agree, then we go to two-port black box circuit representation and we have all combinations of I/O series, parallel, voltage and current feedback. I am afraid that here we stick with just one type and call it “feedback” in general. It is incorrect.
And @Bonsai
My-fi can be pretty pleasing sometimes. 🙂
I sometimes play around with old tube radios, just for fun (plus it's relaxing to work on simple things).
I am always amazed how quickly the masking effect kicks in.
Sure, when you make yourself aware, it's all pretty obvious and audible.
At the same time it's amazing how quickly you get used to it and just enjoy the music.
Sometimes that bit of extra patina on the sound can be satisfying as well.
Very far from proper sound reproduction. Haha
Technically speaking that is true. 🙂
I would call it incomplete instead.
Problem is that every conversation will be an entire article if we have to be 100% correct all the time.
That’s the issue of simplification. And that’s why I do not support high-school textbook like articles. We have plenty of them, it is only about will to learn.
Very nice walk-through on why I should care about slew rates regarding Miller compensation, T.Y.
As for thinking about feedback going "round and round", this only matters in the time domain. On the other hand, thinking of a test signal in terms of frequencies allows parallel processing and multiple iterations. For instance, if an amplifier has a 1kHz sine wave input signal, that can be thought of as an array of DC offsets that can be processed in parallel to produce an array of output voltages. All in zero time.
The output array can then be fed back to an inverting terminal and a new output is calculated, and the slightly non-linear gain causes intermodulation: H2 from the inverting input modifies the gain function for the signal input, creating higher harmonics, and so on with each iteration.
Of course the idea pretends that the gain itself is a true function - it's a model, not a simulation - but it clearly illustrates how the higher harmonics are generated.
But the amplifier is processing instant values of input and divided output to create error control signal. It has no idea what will happen with the input in next few milliseconds or microseconds. It cannot assume for the signal to stay periodic pure sine.
That's not a real engineer. A real engineer loathes the marketing department.
Global feedback was only used in "high end" designs, like the Williamson and Ultralinear topologies. While these amplifiers were firmly established in the market by the 1950s, the overwhelming majority of consumer audio equipment (table radios and phonographs) still use the same ancient topologies that were established by 1940 or so; local feedback only, and the output transformer outside of any feedback loop.
I was literally reading some old magazines from around 1955 or so.
I think half of the schematics of the tube amplifiers in there have global feedback.
(I can post them when I am back home if you really wanna see)
At least in Europe, the majority of tube radios also have some flavor of global feedback. Often in combination with EQ. Incl output transformer and all.
So can you maybe show where you get the information from that only high end brands used global feedback?
Because from all the literature I find, and radios I work on, at least on this side of the pond, feedback was used everywhere in tube amplifiers.
Local, global incl transformers and everything.
First, I have lived in the US my whole life. When I was a kid (in the 1960s) I took everything apart. Even early transistor circuits did not employ global feedback. I took hundreds of table radios and cheap transistor radios apart, reverse engineered them, and put some of them back together - and they worked. That's how I started learning electronics.
Your consumer electronics of the time (Grundig etc) was considered ultra high end on this side of the pond and I didn't even see a Grundig unit until I was about 13 or so. Everything built here was as simple as possible (just like our dinosaur cars).
So my information is strictly anecdotal. Yes I spent many hours studying schematics so I could understand how stuff worked.
Your consumer electronics of the time (Grundig etc) was considered ultra high end on this side of the pond and I didn't even see a Grundig unit until I was about 13 or so. Everything built here was as simple as possible (just like our dinosaur cars).
So my information is strictly anecdotal. Yes I spent many hours studying schematics so I could understand how stuff worked.
One more vintage rare article which describe one simple two stage single-end tube amp which use both negative and positive feedback to reduce output impedance to 0.
(hope I`m not OT)
(hope I`m not OT)
Here is a very cute and interesting example from 1957.
Clearly still with tubes in mindset, but it's one of the first Philips transistor radios, but still with output transformers and such.
But definitely incl feedback
https://www.vintageradio.nl/radio's/philips_l3x71t.htm
Click on schema to get the schematic.
Wouldn't personally consider this as high end. @Fast Eddie D
There are plenty of these simple radios like this. Just have a search 🙂
Clearly still with tubes in mindset, but it's one of the first Philips transistor radios, but still with output transformers and such.
But definitely incl feedback
https://www.vintageradio.nl/radio's/philips_l3x71t.htm
Click on schema to get the schematic.
Wouldn't personally consider this as high end. @Fast Eddie D
There are plenty of these simple radios like this. Just have a search 🙂
And what's considered high-end there is what counts?
I am a bit confused what you're trying to say?
It started originally with a comment from someone that (global) feedback didn't exist in the 50s and 60s.
Which is just not true.
If that's high-end in some parts in the world or not, isn't really relevant.
I am a bit confused what you're trying to say?
It started originally with a comment from someone that (global) feedback didn't exist in the 50s and 60s.
Which is just not true.
If that's high-end in some parts in the world or not, isn't really relevant.
Global feedback certainly did exist in the 1950s and 1960s; example Ultralinear and Williamson topologies were developed in the 1940s and incorporated into consumer equipment in the 1950s.
All I'm trying to point out is that European and American markets were quite different. Look at the difference in automobiles in the 1950s and 1960s; they could hardly be more different.
There is no right or wrong; just different. The perception on this side of the pond was that European consumer electronics was "high end" and in fact it was more technologically advanced than what was produced here.
All I'm trying to point out is that European and American markets were quite different. Look at the difference in automobiles in the 1950s and 1960s; they could hardly be more different.
There is no right or wrong; just different. The perception on this side of the pond was that European consumer electronics was "high end" and in fact it was more technologically advanced than what was produced here.