The strange case of the A class

June 5, 2021
 by Paul McGowan

If you’ve been keeping up with my rant on amplifier classes you’ll recall that in yesterday’s post I explained class AB operation.

One of the takeaways from that explanation would be that without an input signal, a Class AB amplifier constantly draws a small amount of power out of the wall—enough to warm its heatsinks. Not until there is an input signal does the amplifier start to draw much power.

Remember back to the beginning post on efficiency where I pointed out that in a Class AB amplifier, for 100 watts delivered to the loudspeaker another 100 watts would be converted to heat? The net result of that is an increase in heat for every watt delivered to the speaker. More power delivered to the speaker equals more heat generated by the amplifier. Makes perfect sense, right? The harder the amp works the hotter it gets.

Class A amplifiers are the opposite.

In a Class A amplifier, the more watts delivered to the speaker the cooler the amplifier gets! In fact, the point of a Class A amplifier’s maximum power output happens also to be the point of maximum efficiency.

Weird, right?

A 100 watt Class A amplifier with no input signal draws 200 watts out of the AC wall receptacle. All 200 watts are converted to heat. That same 100 watt class A amplifier delivering 100 watts of audio power to the speaker still consumes the exact same 200 watts from the AC wall socket, only this time, half of the 200 watts consumed goes to heat while the other half goes to making music.

Thus, Class A amps are strange beasts indeed.

Subscribe to Paul's Posts

18 comments on “The strange case of the A class”

  1. I recall the early pure Class A amplifier designs from the mid 70s. The first being the Stax Model DA-300 (150w/chnl) and their lower powered siblings DA-80 & DA-80M (45w/chnl). If memory serves well, Stax was then distributed in the US by Dr. Bruce Maier @ the Discwasher Group.

    The DA-300 influenced the soon to follow Mark Levinson ML-2 (25w/chnl) monaural amplifiers and the Threshold 400A (100w/chnl).

    These early solid state Class A amplifiers delivered an intriguing purity of sound lacking traditional transistor sonic harshness (think 70s crown & mcintosh). At their best, a softer silkier sound not unlike Shunyata products but could also on occasion, depending on the recording sound dry, sterile and amusical in the upper frequencies. Compared to the higher power Class A/B offerings of the day such as GAS Ampzilla and SAE 2500 they lacked the warmth, low frequency extension and slam in comparison.

    Over the last 50 years other design engineers have implemented refinements to pure Class A topology yielding greatly improved musical transparency and extension. Ironically, with all the advances in solid state amplifier technology many audiophiles today still prefer the warmth and harmonic richness of vacuum tubes. Just like the great analog vs digital debate, high performance solid state and vacuum tube amplifiers remain expensive.

    Another noteworthy solid state design from this era, though completely different circuit topology were Sony’s VFET amplifiers and the original Class D amplifier dog, the Sony TA-N88. The year was 1976.

    1. as far as i know, Class A was actually the very first operating class in electronics. all single ended tube circuits were inevitably class A (at least for AF. RF is a different matter).

      the invention of push-pull circuits made the more efficient B and C classes possible. but even the earliest p–p amp designs were class A. input or interstage transformers were used for phase splitting, the only real difference to SE was the output transformer which got a center tap in the primary.

  2. The Levinson ML2 was really the purest class A push/pull(since single ended amps must be class A) in that it was class A down to its 2 ohm specs(and maybe lower). Most class A amps are class A into the 8 ohm spec but become less and less class A as the output impedance is lowered. It’s interesting that the ML2 was a low power amp but almost a high current amp.

    1. I was fortunate enough to have access two listen on more than one occasion to the IRSV system driven by an ML-1 preamp and ML-2’s at Lyric HiFi in Manhattan which was a jaw dropping experience. At that point in my audio history, my home system had a Threshold 4000 as part of a system driving Infinity RS1-A’s and found it to be a phenomenal amplifier. Both amps were extremely well-built unlike a prior amplifier that I owned, an Ampzilla which at that time was my favorite but could not compare to the 4000 in any way. To be totally fair, the Ampzilla was driving Dalquist DQ-10’s so it’s not fair for me to evaluate it’s bass attributes. For me those are were my golden years.

      1. Golden, indeed. I recall those times very well.

        If your Dahlquists were the DQ-10A implementation, they’d still be wonders and wonderful today. Someday, I might pick up a pair, if I can find them in great condition.

        Thanks for your informative posts, gentlemen.

  3. Now that we’ve all had the lesson about room heaters. 😀
    What I learned was class D not very good at generating heat
    Class A/B better … want more heat crank it up
    Class A best room heater…. Too much heat? Crank it up.

    One may make the assumption that if one lives in a 4 season environment that at least 3 amps are needed…. D for the summer, A/B for the spring plus autumn, and of course a very quietly played A for those cold winter nights. 😉

    How about a discussion on sound characteristics between (the politically incorrect word) of the classes.

    If an A bias is the gold standard in pre amp circuitry (non high current applications ), then why not all the way thru?
    ….Consumer demand?
    ….Gov’t regulation?
    ….Cost ?
    ….Design requirements?
    ….or a combination of all the above, plus what I may have missed?

  4. As I have mentioned before, there are some interesting corollaries between the physics laws of electronics and hydraulics (Ohms versus Lohms laws).

    Class A amplifiers work very much like a hydraulic system with a constant displacement pump pushing fluid through a pressure relief valve to maintain constant pressure.

    When no fluid is tapped off to perform work (like driving a log splitter or a car lift), fluid passing through the relief valve converts all its energy into heat.

    The most efficient mode of operation for this hydraulic system is to use up as much of the flow available to do work. This minimizes the amount of heat generated.

    But you don’t want to exceed the max flow of the pump capacity with this hydraulic system configuration. Otherwise the pressure relief valve can go unstable, resulting in spikes in pressure (clipping).

    A nice mental image to help understand the operation. At least for me.

  5. “A 100 watt Class A amplifier with no input signal draws 200 watts out of the AC wall receptacle. All 200 watts are converted to heat. That same 100 watt class A amplifier delivering 100 watts of audio power to the speaker still consumes the exact same 200 watts from the AC wall…”
    From what I understand this is just an example.
    IMO it’s good to mention that it’s also possible that a Class A amp draws 400 Watts out of the wall receptable in order to deliver 100 watts to the speakers, depending on the efficiency of the amp.
    In my example it is of course 25%; 75% is converted to heat. On the bright side : with an amp like that you might not need central heating anymore…
    Also the PSU is often more expensive, which also adds to the cost of the machine.

  6. Being strange than me and class A amplifiers should get along just well. 99% of the time I play my system with one or less watts. 5 watt peaks. So getting the first watt right is important to my listening. I’m not interested in amplifiers that have to push out many watts before sounding good. That being said I believe there are good sounding class A, A/B, D, G, H amplifiers out there and some terrible ones. Depends on the parts and how those parts are implemented into the design and the quality of workmanship. Also there are good sounding Mosfet, Fet, Bipolar, and tube amplifiers out there and some terrible sounding ones. None of the designs are bad just because there are some out there that sound bad. Blame the cook.

  7. Ultimately, the amount of heat generated by a Class A amp is constant. After all, the sound output of the speakers just vibrates stuff which then heats up. Eventually all the free energy in the universe becomes heat- hence the phrase, “heat death” of the universe.

  8. ‘A’ class amps are only “strange beasts” if you don’t understand
    how & why they operate the way that they do.
    If you can grasp the fact that ‘A’ class is ‘on maximum’ all the
    time then they are not so strange 🙂

  9. How can what class my amp is?
    I have an Audio Design 10A solid state amp designed by Howard Brown and sold from 1975-1985 from Cambridge Ontario. It’s long defunct and hard to info on it. I really like it and there are lots of online comments on the high quality of this amp. I leave it on all of the time and it doesn’t appear to get too warm.

    Can one tell from its schematic what class it is or does one have to open it up? Here’s a link to the schematic of its bigger sister, the 20A, in case it’s helpful.

  10. Silly me. And here, I though the whole point of class A and AB was to manage the problem of the mismatching of the wave forms when the push-pull transistors or tubes switched from positive to negative versus the reference level and back (this is a poor description trying to use words when a simple diagram would be soooo much clearer) of imperfectly implemented (all of them) class B amplifiers. Of course, all of the class A and AB amps are imperfect, too. I have only had one chance recently to listen at some length to a class A power amp (Gryphon Antilion EVO being burned in by the good old, ever lovin’, brick & mortar stereo specialist for a client; driving a pair of Wilson Alexx speakers). No comparison to any other power amp, but I thought it sounded just fine.

    1. By the way, Mr. Rat, that “at length” listening session with the Gryphon class A amp happened to be of all three sides of Beethoven’s Symphony No. 9 from the Bernstein Centennial box set. Not a steak in sight or smell.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram