"Class A" nonsense

[Merlinb]

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

You, Aiken and I at least appear to be on the same wavelength!

[balijukka]

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".

IMHO class AB amp can not operate class B. There will always be that fraction when the signal passes zero, when both tubes conduct. Very close to class B operation though, so maybe we can say it runs class B.

jukka

[balijukka]

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.

Classy answer, at least we all know what you're talking about.

jukka

[Old Tele man]

...I have a problem attaching the word "audio" with "Class-C" because Class-C is (to my knowledge) only used with RF-amplifiers (most commonly RF-oscillators), whereas, Class-B can be both "Audio" and "RF-modulators."

...my single-tube, "Class-B" amplifier example comes directly from a USN AN/APS-20 Radar (that I used), where the tube is biased OFF (for reduced power consumption and increased tube life/reliability) and only comes ALIVE when a signal is present...and only HALF of the signal contained information of interest, but it had to be faithfully amplified out of the demodulated, down-converted RF/IF-stage signal. Similar circuits were used in the old Tektronic scopes, too, which had bandwidths from DC to 1-5MHz, etc., as I vaguely recall.

...thus, Class-A, -B and -C operation define (as I learned it) 'HOW' the TUBE amplifies the signal, not how 'FAST' it does it (AF vs RF) and it's only the "push-pull" configuration which requires two tubes, ie: there's separate Class-A (SE) and Class-A (PP) operations possible. And, I believe, you'll only find the literature stating that "Class-AB" has the requirement of PP configuration, not specifically Class-B.

...I foresee "flames" and lots of them.

[Steve Conner]

Well, according to my model of things: Nobody in their right mind would design an audio amplifier to be Class-C, as you say. So no audio amplifier should have a design class of C.

However I argue that it's possible to have an operating class of C, by just overdriving the cr@p out of the thing, until you develop enough self-bias from grid current that the tube stays off for more than 180 degrees.

[Merlinb]

...thus, Class-A, -B and -C operation define (as I learned it) 'HOW' the TUBE amplifies the signal, not how 'FAST' it does it (AF vs RF) and it's only the "push-pull" configuration which requires two tubes, ie: there's separate Class-A (SE) and Class-A (PP) operations possible. And, I believe, you'll only find the literature stating that "Class-AB" has the requirement of PP configuration, not specifically Class-B.

Hmm, you present an interesting argument. I'll have to mull this one over some more.

However, why do you exclude class AB? Surely if a single device can be class A (centre biased or hotter), or class B (cut-off biased), then it can also be something in between which would be class AB. According to your system, why does AB have to be PP?

[Old Tele man]

However, why do you exclude class AB?

...that wasn't my call, merely my regurgitation from the literature.

...my 'lightbulb-moment' thought is that it's due to Class-AB having to straddle the two conditions:

Class-A (360º) where Ra = Rpp/2

Class-B (180º) where Ra = Rpp/4

...Class-AB has a "composite" load line, whereas Class-A and Class-B follow a single (not accounting for reactive loads) load line.

...or, it could simply be that was the accepted "convention" and so included in the textbooks of the era 1927-1960s.

[Steve Conner]

That's what I think too. You can't have a composite load-line in a stage that only has one device, because you need more than one thing to make a composition.

[Merlinb]

...my 'lightbulb-moment' thought is that it's due to Class-AB having to straddle the two conditions:

Class-A (360º) where Ra = Rpp/2

Class-B (180º) where Ra = Rpp/4

...Class-AB has a "composite" load line, whereas Class-A and Class-B follow a single (not accounting for reactive loads) load line.

I see what you're trying to say, I just don't see how you can sit on the fence regarding AB. When you said:

...thus, Class-A, -B and -C operation define (as I learned it) 'HOW' the TUBE amplifies the signal, not how 'FAST' it does it

I assumed you we sticking rigidly to the definitions that if conduction is 360 degrees, you have class A.
If conduction is 180 degree, then it's class B.
Conduction more than 180 but less than 360 it much be class AB- regardless of the loading condition, and such conduction could be achieved with a single device.

In fact, the more I think about it, the more I believe that I have never actually read a textbook which says that AB has to be push-pull by definition, only that it that it has be push-pull for low distortion, which is also true for class B of course.

Do you agree?

[Old Tele man]

...thus, Class-A, -B and -C operation define (as I learned it) 'HOW' the TUBE amplifies the signal, not how 'FAST' it does it (AF vs RF)

...that was simply my rather 'poor' way of trying to separate AUDIO (20-20KHz) from RF (radiated E/H fields)...which was technically "incorrect" because electro-magnetic energy extends all the way down to "not zero," but acoustic energy is modulated air-pressure, as opposed to modulated voltage or current.

...good point.

...another point of "...departure..." is how projected-cutoff (almost 100% Class-B) is still called Class-B, but is actually into the Class-AB realm of operation.

[balijukka]

...that wasn't my call, merely my regurgitation from the literature.

...my 'lightbulb-moment' thought is that it's due to Class-AB having to straddle the two conditions:

Class-A (360º) where Ra = Rpp/2

Class-B (180º) where Ra = Rpp/4

...Class-AB has a "composite" load line, whereas Class-A and Class-B follow a single (not accounting for reactive loads) load line.

...or, it could simply be that was the accepted "convention" and so included in the textbooks of the era 1927-1960s.

IMHO the logic goes vice versa.
IF you use class AB operation in push-pull THEN you have composite load lines, due to the other tube stopping conducting.

IMHO the "convention" comes from the fact that most/all textbooks only treat the subject of class AB when dealing with push-pull circuits, it being the only sensible use of class AB.

The class definitions only mention the conducting angles as you have earlier pinted out.

jukka

[Old Tele man]

...most everyone agrees, there's a difference between design Class and the operation class (notice use of lowercase).

...but the crux here is really that ONLY the designers intended Class has meaning (standards, convention, guidelines, definitions from IRE, EIA, REA, JEDEC, IEEE, etc.), because user applications beyond the products intended use are NOT santioned nor supported by the standards, convention, guidelines, definitions of those santioning groups.

...an automotive analogy: a 4x4 Jeep is designed for "off-road" use, so using it for boony-bounching is within its design limits. But, accidently or intentionally climb too steep a path and have it roll over, and the owner/driver has exceeded the vehicles intended design "class" of being a 4x4 "off-road" vehicle. In other words, it was operated outside its intended design "class." Now, doing this did NOT change it's "class"--it's still a 4x4 Jeep--albeit a bent and mangled one.

...likewise: an amplifier designed for Class-A PP operation was created under the design criteria of input signal, bias, and output signal meeting the stipulations of Class-A operation (360º plate conduction). And, like the Jeep, when the owner/player chooses to operate the amp beyond its intended range, it's STILL a Class-A amp--its intended Class of operation hasn't changed--only its actual operation changed. And, this does not change or modify it's orginal design Class-A designation.

...the "sticky-wicket" here is when people go the other way around and claim that "...beyond intended operation..." of Class-A is just an 'extension' of regular Class-A design. Sorry, that's just NOT valid, because any operation that causes less than 360º plate current conduction has violated the Class-A design criteria. Design and intended use Classes are defined; user mis-use are not.

...the three 'tell-tale' characteristics of Class-A PP operation that I look for (yours might be different) are:

1) VOLTAGE - quiescent plate voltage (Vp.q) is about 50-60% of Vbb (B+).
2) LOADLINE - OT Zpp is about TWICE its Class-AB counterpart.
3) POWER - output power is only TWICE its single-tube Class-A rating.

...exceed ANY one of them, and it's actually Class-AB...but, probably, still being called a Class-A amplifer.

[balijukka]

.
1) VOLTAGE - quiescent plate voltage (Vp.q) is about 50-60% of Vbb (B+).

care to clarify this ?
We are talking about p-p stage with an output transformer are we.

jukka

[Old Tele man]

care to clarify this ?
We are talking about p-p stage with an output transformer are we.

jukka

...the quiescent plate voltage (Vp.q) of a single Class-A at idle is roughly about 60% of the supply voltage (Vbb or B+) with triodes and closer to 50% with pentodes.

...Class-A PP is merely two single Class-A tubes operating "inversely" from each other, sharing and operating through a common center-tapped output transformer. Their quiescent and operating characteristics remain unchanged.

...at idle, each plate would be sucking down plate current enough to keep the available plate voltage- and current-swings/excursions "centered" between the plate plate regions of cutoff and control-grid induced clipping (looking like plate voltage saturation). ANALOGY: a single Class-A tube on either end of a "teeter-totter" transformer, one's UP is the other's DOWN, with their their "inversely" continuous Class-A outputs being algebraically summed in the common PP transformer.

...one "tell-tale" sign of Class-AB PP operation is that Vp.q is very close to the power supply voltage (Vbb or B+) for two reasons: (1) most importantly, because they're typically only biased around 195-210º conduction (almost Class-B), and (2) secondly, because the idle currents are MUCH (half or more) lower than for Class-A for the same tube.

[raiken]

...the quiescent plate voltage (Vp.q) of a single Class-A at idle is roughly about 60% of the supply voltage (Vbb or B+) with triodes and closer to 50% with pentodes.

Maybe with a resistive load, but not with a transformer load. In a class A, AB, or B amp, the quiescent plate voltage is equal to the supply voltage minus the IR drop across the primary winding resistance, which isn't very much.