Amplifiers. Class war !

[antibob]

Hi Folks. Can anyone give me, or point me in he direction of an idiots guide to amplifier 'class' types and benefits, disadvantages etc.

I have noticed a few Class A headphone amps recently, I though class A uses a lot of power, I suppose this isnt a problem with a Headphone amp as it wont use as much power for obvious reasons.
I read this info on a manufacturers website, I'd be interested in a short explanation. 'incorporates a discrete class A output stage, whilst the front panel features an additional high quality NKK rotary selector, enabling the end user to select the optimum amount of class A bias for a given headphone impedance.' Is this just techno twaddle or is there some basis of fact.

Just curious about it so no need to go into too much detail, unless you want to.

[Dr Bunsen Honeydew]

Techno-twaddle

Within reason it is all bullshit and used for marketing purposes. The marketing is to produce an amplifier circuit where you have an excuse or a tenuous reason to call it class A. The reason is our perception of quality, like upper class, like 1st class travel, class A amplifiers must be better mustn't they, eh!!

Proper class A's circuits are so inefficient and difficult to use apart from at line level that to use them you have to make compromises elsewhere in the system. Now you can be very happy with those compromises as with LindsayT and his set up which I am sure is excellent (I have only heard his amps and they are definitely excellent).

So we are looking at class warfare where the upper class is always preferable to be even if you have to pretend it. Beyond this stupidity there is common sense 1/ a good item of equipment is a good item of equipment - full stop, no matter if it is class A to Z. 2/ A good piece of equipment will only give its best within a sympathetic synergy.

[antibob]

So is one dsigners class A, another designers class B or are certain circuit designs rearded as A, B etc ?

[Dr Bunsen Honeydew]

Blimey, how long have you got, and why is it important to you?

Class A Amplifiers
Class A amplifiers operate over a relatively small portion of a tube’s plate-current or a transistor’s collector-current range and have continuous plate- or collector-current flow throughout each RF cycle. Their efficiency in converting DC-source-power to RF-output-power is poor. DC source power that is not converted to radio frequency output power is dissipated as heat. However, in compensation, Class A amplifiers have greater input-to-output waveform linearity (lower output-signal distortion) than any other amplifier class. They are most commonly used in small-signal applications where linearity is more important than power efficiency, but also are sometimes used in large-signal applications where the need for extraordinarily high linearity outweighs cost and heat disadvantages associated with poor power efficiency.

Class B Amplifiers
Class B amplifiers have their tube control-grids or transistor bases biased near plate- or collector-current cutoff, causing plate- or collector-current to flow only during approximately 180 degrees of each RF cycle. That causes the DC-source-power to RF-output-power efficiency to be much higher than with Class A amplifiers, but at the cost of severe output cycle waveform distortion. That waveform distortion is greatly reduced in practical designs by using relatively high-Q resonant output “tank” circuits to reconstruct full RF cycles.

The effect is the same in principle as pushing a child in a swing through half-swing-cycles and letting the natural oscillatory characteristics of the swing move the child through the other half-cycles. However, low sine-wave distortion results in either case only if the Q of the oscillatory circuit (the tank circuit or the swing) is sufficiently high. Unless the Q is infinite, which it never can be, the amplitude of one-half cycle will be larger than the other, which is another way of saying there always will be some amount of harmonic energy. (Coupling an antenna system too tightly to the resonant output tank circuit of an amplifier will lower its Q, increasing the percentage of harmonic content in the output.)

Another effective method commonly used to greatly reduce Class B RF amplifier output waveform distortion (harmonic content) is to employ two amplifiers operating in “push-pull” such that one conducts on half-cycles where the other is in plate- or collector-current cutoff. Oscillatory tank circuits are still used in the outputs of Class B push-pull amplifiers to smooth switching transitions from the conduction of one amplifier to the other, and to correct other nonlinearities, but lower-Q tank circuits can be used for given percentages of harmonic content in the output. (Tank circuits can be loaded more-heavily for given percentages of harmonic output where two amplifiers operate in push-pull.)

Class AB Amplifiers
As the designation suggests, Class AB amplifiers are compromises between Class A and Class B operation. They are biased so plate- or collector-current flows less than 360 degrees, but more than 180 degrees, of each RF cycle. Any bias-point between those limits can be used, which provides a continuous selection-range extending from low-distortion, low-efficiency on one end to higher-distortion, higher-efficiency on the other.

Class AB amplifiers are widely used in SSB linear amplifier applications where low-distortion and high power-efficiency tend to both be very important. Push-pull Class AB amplifiers are especially attractive in SSB linear amplifier applications, because the greater linearity resulting from having one amplifier or the other always conducting makes it possible to bias push-pull Class AB amplifiers closer to the Class B end of the AB scale where the power-efficiency is higher. Alternatively, push-pull Class AB amplifiers can be biased far enough toward the highly-linear Class A end of the scale to make broadband operation without resonant tank circuits possible in applications where broadband operation or freedom from tuning is more important than power-efficiency.

Class C Amplifiers
Class C amplifiers are biased well beyond cutoff, so that plate- or collector-current flows less than 180 degrees of each RF cycle. That provides even higher power-efficiency than Class B operation, but with the penalty of even higher input-to-output nonlinearity, making use of relatively high-Q resonant output tank circuits to restore complete RF sine-wave cycles essential. High amplifying-nonlinearity makes them unsuitable to amplify AM, DSB, or SSB signals.

However, most Class C amplifiers can be amplitude-modulated with acceptably low distortion by varying plate- or collector-voltage, because they generally are operated in the region of plate- or collector-saturation so that the RF output voltage is very closely dependent upon instantaneous DC plate- or collector-voltage. They also are commonly used in CW and frequency-shift-keyed radiotelegraph applications and in phase- and frequency-modulated transmitter applications where signal amplitudes remain constant.

I have just copied this across from the web but you can search and find the same or similar. The con men are those that produce Class AB amplifiers and call them Class A because they heavily bias them to A. This is what the recent argument has been all about.

EDIT Sorry just realised this article applies to RF (radio frequency) circuits, but the principles are the same.

[antibob]

I was just thinking you must be an excellent touch typer or have voice recognition on your computer when I read your last paragraph. Most of that was above my head but thanks for going to the trouble of looking it up.

I have also seen the terms 'single ended' and 'fully discreet' is that more bable, a bit like listing the features of a car as having 4 wheels .i.e kind of esential for a car to have 4 wheels ?

[Dr Bunsen Honeydew]

"True" or "Pure" class A is single ended. It means one valve or output transistor is amplifying the whole of a wave form. It is very inefficient which makes for very hot amplifiers and big lecky bills, but low power.

Fully discrete simply means there is no encapsulation of the circuit i.e. use of integrated circuits or chips. The circuit is made up of individual or discrete components.

The problem is technical things are technical, so unless you have the experience of building them you don't understand the nuance, and the nuance is often more important than the gross theory. Just ignore the bullshit and the marketing hype and use your ears. You can't buy by specs nor can you buy by type, if you do you will end up in an endless circle of dissatisfaction - only your ears are important.

[antibob]

Dont worry Doc, that's been my approach shortly after I developed an interest in Hifi as a teenager. I am just interested. I have only ever owned 3 hifi amps, 2 of which are the same brand, I just like how they sound. A dealer once almost lost his temper with me because I didnt want buy a 'bargain' Marantz CD player, but it just didnt do it for me.

[Lindsayt]

There's an interesting side-bar in the Stereophile review of the Krell KSA-250 power amplifier:

Krell KSA-250 power amplifier When is Class-A Class-A?
Sidebar 3: When is Class-A Class-A?

A contentious subject is exactly how much of the KSA-250's power delivery is in pure class-A, which is when both "upper" and "lower" sets of output transistors are turned on all the time. By contrast, class-B is when the upper transistors are turned on only during the time the output waveform lies on the positive side of ground potential and the lower transistors only during the time it lies on the negative side of ground. Class-A/B, which applies to the vast majority of solid-state amplifier designs, involves a small standing bias current to ensure that neither upper nor lower ouput transistors turn off for small signals, thus minimizing crossover distortion.

With amplifiers like the Krell and Threshold models reviewed this month, there will come a point when the output stage's operation changes from class-A to class-A/B, this proportional to the standing bias.

The Krell's rated output of 250W per channel into 8 ohms implies a standing bias current of 3.95A per channel—square root of 250W/(2 x 8 ohms)—if all this power is to be delivered with the output stage running in class-A. Assessing an amplifier's bias current is not straightforward, however, particularly when, like the Krell, it doesn't have a fuse in series with the rail voltages.

Looking inside the KSA-250, the emitters of the 12 pairs per channel of output-stage transistors appear to standing on series resistors of nominal 1 ohm value. The average voltage drop across these resistors was 110.5mV, implying a standing bias for each of 110.5mA; ie, a total of 1.33A. This will give a maximum power for true class-A operation into 8 ohms of 28.5W (14.5dBW) rather than 250W.

Does this matter? I don't think so, as the amplifier's output stage will still run in class-A into 8 ohm loads for all but the final 9.5dB of power delivery, implying that the virtues of class-A operation—constant power-supply stress and constant output-device temperature and transfer function, among other things—will still be there to a large extent.—John Atkinson

Source: http://www.stereophile.com/content/krel ... lass-class

[terrybooth]

Techno-twaddle
Within reason it is all bullshit and used for marketing purposes. The marketing is to produce an amplifier circuit where you have an excuse or a tenuous reason to call it class A. The reason is our perception of quality, like upper class, like 1st class travel, class A amplifiers must be better mustn't they, eh!!

'Nuff said. And as the Doc goes on to point out, if you don't know anything about the detail that this sort of labelling is supposed to represent. Don't bother about it. In fact, ignore it. Trust the force, learn Zen, or just trust the only measure that really matters **your enjoyment whilst listening to it**. Any other enjoyment, like the pleasure of saying that you own such and such or being able to spout endlessly about this that or the other, or that you've spent so little or so much on it will eventually fade into irrelevance.

[Dr Bunsen Honeydew]

Techno-twaddle
Within reason it is all bullshit and used for marketing purposes. The marketing is to produce an amplifier circuit where you have an excuse or a tenuous reason to call it class A. The reason is our perception of quality, like upper class, like 1st class travel, class A amplifiers must be better mustn't they, eh!!

'Nuff said. And as the Doc goes on to point out, if you don't know anything about the detail that this sort of labelling is supposed to represent. Don't bother about it. In fact, ignore it. Trust the force, learn Zen, or just trust the only measure that really matters **your enjoyment whilst listening to it**. Any other enjoyment, like the pleasure of saying that you own such and such or being able to spout endlessly about this that or the other, or that you've spent so little or so much on it will eventually fade into irrelevance.

But the problem is not for a certain idiot who runs a hi-fi forum - for him what he says about his system is to convince people it is the best - so by association he is the best to choose it - is what is *ALL* important. It can't be left to "wow my system sounds good with this (or by doing this etc)" it has to be hyped and bullshitted to the heavens, with nonsensical technobabble, just repeated from his "selected" suppliers (who of course must be the best because *HE* chose them). For me this problem is taking over from the hype and bullshit we had to put up with from magazines and reviewers for so many decades - it screws the industry and our enjoyment away from reality.

The only thing we can say is we like what we have, but maybe there is better. It is impossible to say we have the best as we cannot hear everything, but we can say we love what we have got. So the important thing is enjoy your system and music but keep an open mind, and don't let these idiots pollute your enjoyment.