Class A in push-pull vs Class A/B, etc.

[Free]

I believe I know this stuff, but I'm not positive, so I was hoping for confirmation. It's my understanding that true Class A designs indicate the following: Cathode-bias, non-negative feedback, as well as, single-ended output design. What confuses me is that the Class A tag used by many is actually Class A/B in cathode-bias with no negative feedback. Is this right? I mean it's my understanding that Class A/B means push-pull - is that right? It seems that the vast majority of the 18-20 watt class amps are all Class A in every respect except the push-pull output stage. Also, I thought the maximum two 6V6 tubes could get in true Class A is about 12 watts? Thanks for the help!

-Mike

[Weathered]

I believe I know this stuff, but I'm not positive, so I was hoping for confirmation. It's my understanding that true Class A designs indicate the following: Cathode-bias, non-negative feedback, as well as, single-ended output design. What confuses me is that the Class A tag used by many is Class A/B in cathode-bias with no negative feedback. Is this right? I mean it's my understanding that Class A/B means push-pull, when the vast majority of the 20 watt class amps are all Class A in every respect except the push-pull output. I thought the maximum two 6V6 tubes could get in true Class A is about 12 watts btw. Thanks for the help!

-Mike

Class A can actually be either fixed or cathode biased (IIRC) and can be set up in SE, Parallel SE, or PP configurations. The majority of amps that are marketed as Class A are not actually operating in Class A, rather they are cathode biased, EL84 based amps with no NFB, much like the AC30. I'm not sure where the AC30 being Class A myth came from, but the AC30 operates in Class AB, albiet being biased relatively hot.

The operating class simply signifies how the output tubes work - it doesn't imply anything as far as tone goes, nor does it make one amp superior to another. If an amp works as Class A/B, it has to be push-pull, however, but a push-pull output section is not exclusive to Class A/B, if that makes sense. Some engineer decided one day that Class A was the new "Point-To-Point" catchphrase, and it's been abused ever since.

Check out the Aiken Amps website - they have a good primer for operating class there that is much more in depth than I can get.

[Free]

Thank you, Wheathered. I'm off to the Aiken site right now...

[Wakarusa]

This question comes up fairly often and the answer is always the same: class of operation does not rely on output geometry (whether single ended, push-pull, or parallel) and does not rely on bias scheme (fixed, cathode, etc.).

Instead it depends on how much of the input cycle appears in the output of a power tube. I'm wondering if statements like "conducts for more than 180 degrees of input but less than 360 degrees" (i.e. Class AB) make sense to everyone. Hopefully, this will help...

http://www.wakarusaamp.com/tmp_pix/tgp/sinclassgraph.jpg



Each line represents what the output would look like if the input is a complete sine wave. For class B and class AB only one tube in the push-pull pair is shown.

So, for class A, the tube is "on" for the entire time and reproduces the full sine wave. If there's more than one tube then all of them conduct like this and the results are summed in the output transformer.

For class B only half (180 degrees) of the sign wave appears at the tube's output. For class AB it's a bit more than half. Both B and AB scheme's assume that there's another tube somewhere working with an inverted version of the input signal (e.g. the output from the other side of a phase inverter) so the "other" tube will take care of the "other half" of the sine wave. The two halves (or the two "a bit more than halves") are again glued back together in the output transformer with correct phasing so you ultimately get a sine wave at the output transformer secondary.

[Wakarusa]

In other words, is a given tube amp Class A by necessity when the output transformer is configured for single ended operation?

Short answer: Yes

Long answer: I think a better way to say it is that single-ended doesn't create class A operation, but instead requires class A operation for a musical instrument amplifier. Class of operation is a question of bias so you could bias a single-ended output section to something other than class A. It would just sound like poo because a chunk of the input signal would go missing in the final output.

[John Phillips]

Short answer: Yes

Long answer: I think a better way to say it is that single-ended doesn't create class A operation, but instead requires class A operation for a musical instrument amplifier. Class of operation is a question of bias so you could bias a single-ended output section to something other than class A. It would just sound like poo because a chunk of the input signal would go missing in the final output.Actually most single-ended guitar amps are biased like this (usually too far to the forward end), and do clip quite asymetrically. I agree that sometimes this sounds quite poor, although some people seem to like it :). It seems worse in some amps (eg SF Fender Champs) than others (eg Tweed Champs) - I'm not sure if it's due to the amount of asymetry, or something else. SF Champs definitely sound much better biased cooler so they're much closer to symetrical, IMO.

Whether you should call these asymetric single-endeds Class A (defining full power as the point where the first end of the waveform clips), or Class AB (defining full power as the point where both ends of the waveform clip), I'm not sure...