"Class A" nonsense

[Robert M. Martinelli]

First you come out preaching that IEA rules say this and that, then you could not show these IEA rules, and now you're quoting established convention. Soon it'll be some holy script that only the annointed are able to understand.
Somehow this reminds me of the church in Rome trying to silence poor Galilei.



The amps are like the real world, not black and white. There is a very big grey zone between the idle and maximum output, and it happens that it is the one most used.

Individual thinking and free expression are the roads to enlightment, but as all the preachers know, if you keep shouting out loud, very few people would bother to raise a voice. And perhapr even the dead horses will resurrect.

jukka

Oh my!
Sorry, I couldn't keep myself from commenting about this, even if I said I was not willing to participate any longer.

With all due respect :

First of all, literally HUNDREDS of textbooks ( and I' ve been reading/studying quite some of them, both in Italian and in English ) with NO exceptions, describe the classes in operation in the same way we did before, so we're not talking about ( more or less useful ) conventions.

Secondly, talking 'bout "poor Galilei" ( and, living closer to Pisa, Italy than you, I think I know something more about this matter ), who' s playing the "church's inquisition" role? Did anyone here behave like Torquemada? Did someone "silence" you or threaten you to burn you on a stake? It seems to me you've been left absolutely free to express your ideas so far, and everyone has behaved correctly IMHO...And....most importantly, who's playing Galileo? You? Do you think you really have something as new/as important to offer to the world? I know my limits and know I can't apply for a Nobel prize, but, if you feel you're up to the challenge, and want to enlighten the world with your knowledge, be my guest....

Thirdly, you keep on talking about amps being everything but black and white, and in this respect I can agree, you can use them at any volume between "0" and "10" ( or "11" if you happen to be Nigel Tufnel" ), but, just like my previous "engine" analogy you found so unfit ( I'd have bet on that! ), their rated power is given AT A POINT, ( what do you think the @ "at" symbol you see on specs and datasheets stands for ? it states that the measurements are taken @ that point ) and not all over their usable range.

If we want to compare apples with apples we NEED to agree that the class of operation of an amp, ( any amp, guitar, bass, Hi-Fi, RF ), has to be stated at one ( and one only ) point, which, so far, and rightfully IMHO, happens to be the full un-clipped output. If this to you is talking about "black and white"....well, it is...it HAS to be this way.

Or....are you going to write a new textbook to replace the ones currently used in thousands of universities?

Come on.......give us a break......

Feel free to go on endlessly, if you like - keep on shredding this dead horse until the pieces are of sub-molecular size, but this will be my last post ( this time for good ) about this matter.

Peace, and great tone to all!

Best regards

Bob

[balijukka]

First of all, literally HUNDREDS of textbooks ( and I' ve been reading/studying quite some of them, both in Italian and in English ) with NO exceptions, describe the classes in operation in the same way we did before, so we're not talking about ( more or less useful ) conventions.
Bob

Please show me only one of those hundreds that says that the measurements for class of operation are to be taken at full unclipped output, and I will say no more.
I have pointed out 1938 IRE definition, which is shared in RHD4, that clearly says that the bias voltage AND THE AC VOLTAGE AT THE GRID are set so that the current flows.....

Thirdly, you keep on talking about amps being everything but black and white, and in this respect I can agree, you can use them at any volume between "0" and "10" ( or "11" if you happen to be Nigel Tufnel" ), but, just like my previous "engine" analogy you found so unfit ( I'd have bet on that! ), their rated power is given AT A POINT, ( what do you think the @ "at" symbol you see on specs and datasheets stands for ? it states that the measurements are taken @ that point ) and not all over their usable range.
Bob

I agree with you about the rated power measurements, but none of these data sheets (at least the hundreds, that I have read) states that the class definition current measurement is to be made at maximum unclipped output.

If we want to compare apples with apples we NEED to agree that the class of operation of an amp, ( any amp, guitar, bass, Hi-Fi, RF ), has to be stated at one ( and one only ) point, which, so far, and rightfully IMHO, happens to be the full un-clipped output. If this to you is talking about "black and white"....well, it is...it HAS to be this way.
Bob

The class of operation of an amp is instantaneous dynamic character.
That's why it's called class of operation, and that's how it is defined in the books. Just like that poor old horse before it died, it could run, and it could walk. When he was walking, he was a walking horse, and when he was running he was a running horse. Different classes of operation.

If we want to talk about class of an amplifier, then the definition is open, at least I have not seen one, but I agree with you 100% about the full unclipped output. It's the only one that makes any sense.



[QUOTE=Robert M. Martinelli;123828]
Or....are you going to write a new textbook to replace the ones currently used in thousands of universities?
[QUOTE=Robert M. Martinelli;123828]

Nah, I think I'll write about the adventures of the dead horse.

Or who knows, maybe they'll use in the universities.

jukka

[Old Tele man]

...preaching that IEA rules...these IEA rules...

...uh, it's Electronic Industries Association (EIA).

...and, while I haven't been annointed, I have been whacked upside the head a couple times.

...I'll concede de-feet and simply agree 100% to disagree with you (ha,ha)!

[balijukka]

Keep it civil please folks

Thank you Sir.

I apologize my bad behaviour.

jukka

[R.G.]

I grew up in the edges of the Ozark Mountains, a very backwards place in the USA. Nevertheless, some people there developed great wisdom, in spite of a lack of formal schooling. Some bits of wisdom were passed along as sayings about pigs, which were very common as a food source, and it was presumed that everyone knew something about the nature of pigs and could understand the sayings in reference to pigs.

The particular one which comes to mind is this:

Never wrestle in the mud with a pig. You get dirty and tired, and after a while you realize that the pig enjoys it.

[Old Tele man]

Will Rogers certainly was a VERY astute observer of american culture, at both sub- and supra- levels!

...I recognized a Devil's Advocate, when I saw/heard/read one...because *I'm* one too (a mirror can be a very enlightening thing when you're contemplating...).

[Alex/Tubewonder]

Jukka ,

Here is a simple challenge for you: please present a linear amp design that none of us designers will be able to identify as belonging to any of the classes as defined by convention i.e. A,B, or AB, or any of the sub classes.
By a linear design I mean an amplifier that preserves signal continuity, so PWM amps don't qualify.
Myself, I'm happy with amplifier class definitions as they are.
Galileo & Co has proven to be right, can you?

[Steve Conner]

Here is what Douglas Self has to say about amplifier classes:
Audio Power Amplifier Design Handbook - Google Books

To my mind, the class definitions are by no means standard. I agree that rated power output is the point where the amplifier's design class should be honoured. For instance, in a Class-AB1 valve amp, no grid current will flow at rated power, but in a Class-AB2 amp, the rated power can't be achieved without grid current.

Note that rated power is not always unclipped output. It almost always is in audio amps (excepting cheap car stereo ones!) but to preserve the generality of the class concept, we must consider all classes of amplifier. A Class-C or Class-E amp has its operating point well into saturation, so the rated power is considerably more than the unclipped power.

Going back to that Class-AB2 amp, if we turned it down, so that the power was reduced and grid current no longer flowed, it would still be a Class-AB2 amp, for sure, because that was what it was designed to be. But I (and Douglas Self and many others) would say that, since there is no longer any grid current, it was now "operating in Class-AB1." This is a tricky point, but I believe it to be a valid one. And likewise, we can say that a push-pull Class-AB amp "operates in Class-A" at power levels low enough that both tubes are always conducting, even though its design class remains AB.

To keep nit-pickers in this thread happy, perhaps we should make a distinction between design class and operating class, the latter being not really a "class" at all, just a convenient shorthand, where we describe the amp's operating condition at that moment by analogy. It's the design class of a hypothetical amp whose design operating conditions are analogous to the actual amp's current conditions.

So hopefully we can all agree that the Vox AC30 is a Class-AB amp, while still appreciating that its operating "class" runs the whole gauntlet from C to B to AB to A with the decay of every note.

[Old Tele man]

...perhaps we should make a distinction between design class and operating class...

...intended use and actual use may vary...

...great minds think alike (sometimes)!

[balijukka]

By a linear design I mean an amplifier that preserves signal continuity

That's the key.

As I said before, going semantic, we can speculate that the word amplifier means that it only amplifies. It does not clip or do any other funny things to the signal.

If it is agreed, there is no problem at all.
It all becomes very clear and the amplifier class definitions become mutually exclusive.

If an amp operates only in class A, it is a class A amp.
If it can operate class B, it's a class B amp. (it can not operate class A or AB)
If it can operate in class AB, it is class AB amp, since neither class A or class B can not.

It also means that single valve amp can only be class A, and class AB and class B must have minimum two valves, as Merlin pointed out. (I think we can forget the one frequency tuned RF amps here, them not being linear)

It also means that there is no need for the maximum output requirement, perhaps that's why it was not included in the old definitions.

jukka

[Old Tele man]

...a single tube can be operated either Class-A or Class-B...either conducting all the time (-A at 360°), or sitting at zero plate current until a signal becomes present to cause the tube to unilaterally pulse an output (-B at 90°). Why you'd want an amplified HALF sinusoid would depend upon the intended use, but it is a valid application example, and it would be Class-B linear (basically a variable amplitude halfwave rectifier).

...and, there is a similar NON-LINEAR example--the "old" squarewave generators worked this way...you fed in a sinewave, the tube conducted only on the positive pulses (which brought the control grid 'out' of cutoff) to produce a fairly "sharp-sided" squarewave output at the plate, which bascially swung back and forth between cutoff and saturation...obviously NOT linear, because the definition of a squarewave is the infinite sum of odd harmonics, added in-phase. Overdriven Class-B...no signal, no plate current.

...and, two tubes can be Class-A , Class-B, or Class-AB (which, by definition can only be push-pull)...but, NOT to be confused with "parallel" two-tube operation.

[mooreamps]

It also means that single valve amp can only be class A, and class AB and class B must have minimum two valves, as Merlin pointed out.

Ole Teleman has it right. Maybe this guy "merlin" needs to do do more than just write another book......


Perhaps not the best forum to discuss electrical theroy, since even some of these so called "gurus" can't even get it right. I've thought about writing a book or two myself on the subject, but it seems to me KOC has it already covered pretty good. The only thing I could write about is the extra stuff I do, and I'm not so sure I want to do that.

[balijukka]

Ole Teleman has it right.

old Tele Man has it right, but he answers to the wrong questions.

What Alex was talkin about was linear application.
I answered that if that unclipped output signal is a premise, it'a all clear, etc. etc.

If you want non clipped linear output, this class-B linear, falls outside the premises.

If not, then that's another matter.
I think everyone who has worked with radio equipment knows that they have class B single valve circuits.

jukka

[Merlinb]

Ole Teleman has it right. Maybe this guy "merlin" needs to do do more than just write another book......

I'm not sure why you have taken a dislike to me in particular, Gary, but I think we are running into a sort of grey area between AF and RF here. Here's what OldTeleman said:

...a single tube can be operated either Class-A or Class-B...either conducting all the time (-A at 360°), or sitting at zero plate current until a signal becomes present to cause the tube to unilaterally pulse an output (-B at 90°). Why you'd want an amplified HALF sinusoid would depend upon the intended use, but it is a valid application example, and it would be Class-B linear (basically a variable amplitude halfwave rectifier).

Now, everyone knows that you can bias a single valve at cut-off (or beyond) and you get a half-wave rectifier.

The question then is, is it an audio amplifier?

Obviously you can feed in an audio signal, and obtain a half-wave rectified signal which, although being monstrously distorted, is still "sound"; it is audible.

Now, my viewpoint is that if this amplifier is not designed for audio use then you would be right to call it a class B or C amplifier.

However, if it is designed for audio use then what you have is a Class A amp. A really badly designed class A amp.
Before you jump in, let me clarify what I mean:

I have already said in another post that you can design "a class AB amp" but observe that at different signal levels it can be said to be "operating in class A", or "operating in class B". But, overall, the thing standing in the corner is a class AB amp. Even when it's switched off. That's what it's called, because on average it combines class A and B operation. It is not a "class A and B amp" (although class A+B valve amps have been designed, but that's another story).

If you now take a single valve set up for pure, honest-to-goodness class A, you call it a class A amp. If you heavily overdrive it so it hits cut-off then you might say it is 'operating in class AB', but the thing in the corner is still a class A amp. It doesn't "become a class B amp", it is an overdriven class A amp, some of the time.

Now, you increase the bias voltage, making it colder. You still have a class A amp, all you are doing is altering its input sensitivity, that is, reducing the unclipped output power. Nothing else has really changed.

If we take this to the extreme we can bias it at cut-off. We now have a class A amp with infinitely small input sensitivity. That is to say, up until the point of clipping it is conducting for 360 degrees. The problem is, the point of clipping is immediate.

That is why I am happy to say that a single valve can only be described as class A for audio, even though it may be operating in class B/C some or even all of the time.

I await flames...

[Steve Conner]

I await flames...

Not from me, I agree completely.

When you say "This is a Class-AB amplifier" you're talking about its design class, and when you say "The stage is operating in Class-A" you're talking about its operating class.

They're two different classes of class, if you like.