Class A or A/B amps

Discussion in 'Audio Hardware' started by motorcitydave, Jun 23, 2012.

  1. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Which do you prefer and why? :cool:
  2. Doug Sclar

    Doug Sclar Forum Legend

    Location:
    The OC
    Class A amps by far.

    They are much less effecient, but don't have crossover distortion which AB amps have. The devices in a Class A amplifier are biased in a manner so that they are always operating in their linear region.

    When you take a standard transistor and apply voltage to it is doesn't turn on until it gets to a certain point, usually around .6v. To operate in Class A, the transisitor is always turned on, so that that non responsive region is taken out of play. Of course this means that the idle current will be substantially higher than that of an AB amp.

    Of course this is an oversimplification and just one of the advantages of running in Class A mode.
  3. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Cool, thanks for the explanation. I like the sound of Class A myself, even though I own an A/B amp. :D
  4. KT88

    KT88 Forum Resident

    As with almost everything, you can't reduce it to such a simple choice. Doing so will remove many superior products in either category from consideration. That is because there are excellent designs of both topologies to choose from. There are also some very poor designs in both categories.

    They can be very different sounding and the largest factors will be cost and required power. If you have a fair amount of money and efficient speakers, you can choose Class A. If you have a lot of money and inefficient speakers, you can still choose Class A but the amps will be enormous. Due to their inefficiency, Class A designs cost a lot more per watt of output than do Class AB amps.

    There are many excellent Class AB amps and many have significantly lower levels of distortion and better overall load behavior than some Class A types. If you were to ask the same question to an amp mfr who produces both, you'd almost certainly get the same conditional answer. It depends upon your power needs and budget. If there were the unusual example of two amps that had equal build quality and cost, yet one was Class A and one was Class AB, the advantage of the Class AB amp would be that it would be the most flexible of the two in terms of driving future speakers and it would have the highest amount of headroom in any case. That said, if you only need 50w max and you can pay for the top quality build required to make a reliable Class A amp, then it is a great sounding choice.

    Many tube amplifiers are Class A and just like their solid state counterparts, they are lower powered and can be rather expensive on a per watt basis. As the first watt or so can be very sweet, they can be very pleasing to listen through. The lack of headroom however can result in compression and a colored sound which is softer than the original signal.
    -Bill
  5. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Some Class A amps like the Pass Labs and Accuphase (and I believe Musical Fidelity and Plinius) double down as the impedance gets lower. So, I don't see a problem with headroom or dynamics with those amps.
  6. McGruder

    McGruder Well-Known Member

    Location:
    Maryland
    Class A/B has gotten better, and there are several really excellent ones that are biased in class A at moderate power levels. I general, I think good class A sounds better than A/B. I love a good class A (ex Pass Labs XA series), but don't like the inefficiency and the excessive heat they give off.
  7. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Can Class A solid state amps shut down thermally if driven too hard?
  8. KT88

    KT88 Forum Resident

    I've owned a lot of both types and sold even more. It is never that simple. You are correct that some Class A amps can double down their power spec. So can many Class AB designs. That is a function of the power supply for the most part, you cannot exceed that and the trouble then becomes the inefficiency of the Class A design. In order to realize a significant power output in Class A, the power supply must be massive. That costs a ton of money. Also with that sort of power delivery at relatively lower wattage than Class AB, you are looking at significant current swings. That means tons of paralleled output devices and massive heat sink area, again that also costs a ton of money. So what happens when you add-up two tons of money? You can afford a pair of bad-*** Class A amps like the Pass XA100.5, which at $16,500.00 a pair, produce only 100wpc. Now, that's all the power that I'd ever use but it isn't always enough for some users, especially those who fancy those planar magnetic things... :winkgrin:
    -Bill
    :cheers:
  9. Metralla

    Metralla Joined Jan 13, 2002

    Location:
    San Jose, CA
    Although I own mostly Class A amps, Dan Agostino's Momentum power amps run in Class AB. They sounded phenomenal driving Wilson Sasha Watt Puppy speakers at last year's Cal Audio Show.
  10. KT88

    KT88 Forum Resident


    The short answer is, yes. It can be thermally protected but it can also cause nuisance tripping as it goes against the principle of the thing. I suppose the greater concern is if the thermal protection will actually be effective in the long run as if the amp were designed to limit the heat produced, it'd probably not be biased as hot, resulting in a sound closer to Class AB and thus defeating the purpose.

    Class A amps actually are more efficient at full output. Only when they are really underbuilt will they go thermal. A lot of designs can literally bake themselves into short lifespans from sitting powered at idle on a shelf. It is just so expensive to properly heat sink them that many mfrs do not. They only have to warranty them for a few years and you better believe that the hotter amps will not outlast the cooler ones, all else being equal. Again, all else is never equal, but hopefully you get the picture there.
    -Bill
  11. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Yeah, I asked that because my old FM Acoustics 611 (Class A) amp used to go into thermal shutdown if driven real hard. Crazy that it did that considering the price i paid...
  12. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    I see your point, but the sound of Class A can't be beat, or can it?
  13. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Yes, I've heard the D'Agostino amps at CES. And yes, they sounded great!
  14. Steve Hoffman

    Steve Hoffman Audiophile Mastering Your Host

    Location:
    Los Angeles
    Good morning, y'all.

    Class A or A/B? How about Class B! Yeah, amazing sounding old McIntosh like MC30s, etc.

    It's all in the design. I've never really let something like Class A, B or whatever influence my decision.

    In GUITAR AMPS however, I love a pure Class A like a Vox. Different thing though.
  15. Get a Plinius amp that has a button on the front to switch between class A and class AB operation. Perfect for A/B testing class A and class AB sound.
  16. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Ok Steve, sincere thanks for your infinite wisdom! :cool::thumbsup:
  17. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Not a bad idea.
  18. KT88

    KT88 Forum Resident

    I'll just let you reflect on the first part of my initial reply then:

    I love the sound of a good Class A amp. I also love the sound of a good Class AB amp. Poorer examples of either are disappointing in their own ways, so there is no clear advantage to simply being Class A in my experience.

    Lastly you really would benefit by considering the load as an integral part of the amplifier.
    -Bill
  19. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Ok, I see your point. Thanks.
  20. coffeecupman

    coffeecupman Forum Resident

    Location:
    Austin, Texas, USA
    Class A in the winter when you have to heat your house anyway.

    Class A/B when you are using the AC for the rest of the year.

    No sense paying to heat a room just to pay to cool it off again.

    My FM111 monos, despite being made by the same company as your thermally challenged 611, are too cool for school. Perfect for hot Texas climes.

    ccm
  21. motorcitydave

    motorcitydave Enlightened Rogue

    Location:
    Las Vegas, NV, USA
    Yeah, I figured that. That's cool you have FM Acoustics, great amps, and the sound is very good indeed. I think the 611 just didn't have enough heatsinking to handle high volumes and as a result high temps, hence the overheating. Just a wild guess.
  22. Burt

    Burt Well-Known Member

    Location:
    Kirkwood, MO
    True Class A amplifiers do not get hotter when driven hard. They are as hot quiescent as they are going to get.

    The Class A Power Point of an amplifier is by definition lower than one half of its quiescent dissipation.

    That said, there are very few push pull audio power amplifiers that operate only in true Class A.

    It is very important to be able to plot a load line vs. the characteristic curves of any device, to be able to discuss this in any detail.

    The current "doctrine" of solid state power amplifier design, as set forth by Douglas Self and his acolytes, both professional and amateur (e.g. the late Randy Slone), is that a power amplifier should operate either in pure Class A or Class B, i.e. with substantially no quiescent dissipation, as operation in Class AB (no slash here folks) results in increased distortion due to a mechanism he calls transconductance doubling , as well as higher power dissipation.

    Douglas is not a nice man (see his correspondence with Norman Thagard in AudioXPress, circa 2004), but what he says makes perfect sense if you assume a power amplifier is a constant power output device. In some kinds of service, perhaps sound reinforcement at death metal concerts, or driving shake tables (but I repeat myself) that is probably not far wrong. In hi-fi service there is nothing farther from the truth, which forms the basis for two radically divergent arguments, the idea that because overload is inevitable only the first watt really matters so get a single ended triode, and that for ideal use get so much power overload isn't inevitable, such as the 1 kW McIntosh monoblocks with Power Goose just in case.

    I take the middle ground: Build an amplifier that runs in Class A for most of its time and seamlessly transitions to Class AB when needed. The tiny increase in THD from gM doubling beats hard clipping by a huge amount, and without Power Goose that's what overdriving a solid state amp does.
  23. Dave, check this one out - XTZ's A100 D3 amp. 110wpc, but the first 50 are output in Class A. Well reviewed (as is the same company CD player). Similar if not identical to the Harman HK990 (which doesn't offer Class A).

    The XTZ comes in at around £600 in the UK, but that's buying direct - so there's no dealer or distributor margin to factor in. Consequently, with that in mind, I think this amp would come in somewhere around the £1200 mark or so. That's incidental, but it's a fine bit of kit.
    darkmatter likes this.
  24. coopmv

    coopmv Newton 1/30/2001 - 8/31/2011

    Location:
    CT, USA
    It appears XTZ amps are actually available in the US as well ...
  25. Burt

    Burt Well-Known Member

    Location:
    Kirkwood, MO
    I love the AC30, but it actually has a Class A power point of about half its actual power output.

    This page has a great explanation. I would cache it for those interested since it appears Mr. Aiken's home page shows evidence it might go away.

    http://aikenamps.com/VoxAC30classA_2.html

    I will add that I have never seen an AC30 that didn't put out a lot more than 30 watts into a noninductive dummy load. It is realistically a 40 watt amp more or less, depending on the tubes, the line voltage and the cathode resistor. And because its speakers are light and efficient it is loud.

    Is the McIntosh unity coupled amplifier in its classic iteration (all the tube power amps besides the MI200, MC3500 and MC225) a Class AB or B amplifier? Good question. Best answer so far is that it operates in a regime that doesn't fit either description perfectly.