Class of operation for Tube Power Amps

As mentioned above, SET is pure class A "by definition", though you can have SET with paralleled output tubes which - from the outside - is hard to distinguish from a push-pull configuration. 

Most tube amps are push-pull AB. The reality is that running power tubes in full class A necessitates either drastically shortening tube life or cutting power output way, way down. Neither of which is conducive to sales. And on the flip side, though I have no hard evidence - if "feels" to me that most of the sonic compromise attributed to Class AB solid state (versus Class A SS) is mitigated in the PP tube domain - maybe due to the larger amounts of "benign" low order distortion (masking effect) and/or lower amounds of GNF - contributing to masking the AB crossover distortion (or whatever).

Cathode bias circuits used to be quite common - these tube amps don’t have normal bias adjustments - maybe a "balance" pot at most (to help match the push/pull sides). These amps run hotter per Watt, and much further into class A territory. They’re not so common now; auto-bias circuits (Prima Luna, VAC, ARC) are the new way to achieve cooler running combined with "optimized" performance, and old school cathode bias eats up modern Russian / Chinese power tubes. It was a different story back when we had golden-era tubes aplenty - Mullards / RCA / Tung-Sol / Sylvania / etc - that you could abuse all day (every day) and they’d keep ticking for years. Lots of Mullard power tubes out there with visible horrid scorch markings on their plates, that still measure & play fine lol. 

Tube pres and phono stages are pretty much all class A - power is not an issue there. There’s no reason not to do class A. 

dynamiclinearity

In fact any push pull amp solid state or tube is unlikely to be class A at very high powers although it may be what's called rich class AB . . .

My former monoblocks, CR Developments Artemis, were P/P Class A - all the way.

@steakster what do you mean by class A all the way? The amplifiers may actually have higher power in class A/B above what they are rated for in the spec sheet at class A.

just plug amp into ac power (or current) measuring unit, and dial the volume control.. class A power consumption is not sensitive to volume, while AB will consume significantly more with volume increase. 

@westcoastaudiophile This is only true if the amp is class A1. If A2 or A3 things might be different. 

Look at the number of output tubes: If the amp has only one or two output tubes, it's more likely to be Class A. If it has four or more, it's likely Class AB. 

This statement is misleading. We make tube amps with 20 power tubes and they are class A. Whether its class A or not depends on the operating point of the tube (plate Voltage and plate current) not the number of tubes!

Cathode bias circuits used to be quite common - these tube amps don’t have normal bias adjustments - maybe a "balance" pot at most (to help match the push/pull sides). These amps run hotter per Watt, and much further into class A territory. They’re not so common now; auto-bias circuits (Prima Luna, VAC, ARC) are the new way to achieve cooler running combined with "optimized" performance, and old school cathode bias eats up modern Russian / Chinese power tubes. It was a different story back when we had golden-era tubes aplenty - Mullards / RCA / Tung-Sol / Sylvania / etc - that you could abuse all day (every day) and they’d keep ticking for years.

@mulveling Just so you know this statement is 100% false. You can cathode bias a pair of power tubes to be AB and running a power tube cathode biased does not affect its life in class A, so long as the tube is operated within its specified parameters. So those tubes you're referring to above 'kept ticking' because they were not being abused.