[Tekker]

I was just writing a reply for Any S about his tube amp book in the "Cool Amp Company" thread when the thread disappeared... So I'll start a new one for it.

Originally Posted by Andy S

So it all was making sense....
except the "near class A" amp. I have an idea what they're trying to say, but it's either class A or class AB that's biased to behave like A. (or so I THINK that's what the book said )

I'm not sure if you meant class B or class AB behaves like A, because class A IS A. But yes, class AB behaves like A in that there are no "gaps" in the signal like class B has.

Class A uses a single tube (or transistor if it's a solid state amp) for both the positive and negative parts of the signal. So it uses half of its available voltage for each side.

Class B and class AB both use two tubes (or transistors), one for the positive half of the signal and the other for the negative half. For transistor takes a certain amount of input voltage to "turn on", so part of the signal is wasted just to turn the transistor on. If you have to turn two of them (for positive and negative) you have two transistors to turn on and you loose twice the amount. This is the problem with class B amps. Class AB however is designed to add some extra voltage to the input of the transistors so that they are always "on" and the signal remains continuous without any gaps.

I don't know the benefits of using class AB with tubes however, because I don't think they require input voltage to turn on (I don't know much about tubes though, so I could be wrong). I have a tube amp book also, but I haven't read very far into it... So I guess I better make my way through it.

-tkr

[allthumbs]

Sorry tekker. The amp company post was posted by the owner of the company which made it self promotion so it was deleted.

[P-90]

Well, since this topic has a whole thread to itself now, I'll give it a go.

Class A can be implemented in single devices (tube, transistor), or in pairs/sets of devices. Let's assume you have a transistor which can go between 0v and 20 v. For Class A operation, you set it to run with an electrical center of approximately +10 volts, so positive swings of the wave can go up 10 volts, and negative swings down 10 volts. Because you're always running 10v of DC through the device, it will run very hot; even when you're not playing a note, it is having to work at 50% of its capability, just passing that DC through. Class A tends to produce the least distortion, but to use the most power. This also means that much of your power is NOT going to be available for amplifying the signal, it's just going to run through the device as DC, and be dissipated as heat. It is almost universally used in preamplification stages, but more rarely in output stages. It is often used in very small (10 watts or less) guitar tube amps, but otherwise is found mostly in preamp stages and in the output stages of posh hi-fi equipment.

Class B is used with pairs (or sets) of output stage devices. To keep from heating the house with the wasted electricity of a Class A power amp, one uses one device to handle the positive half of the wave, and another to handle the negative half. To continue the 20 volt example, for Class B, one transistor would produce the output between just over 0v and +10v, the other would handle everything below 0, as far as -10. When there is no signal, rather than working hard like a Class A amp, neither transistor has to work at all, they are both shut off, since the signal's neither positive or negative. When the signal does go positive or negative, it takes a tiny bit of time for the transistor to catch on and start amplifying again, and as it approaches (or comes out of) shutoff, it tends to act pretty nonlinear, i.e. it has major distortion problems. The turn-on lag and nonlinearity are known as as "crossover distortion." It's not too bad a problem if you have the volume up high, but if you have a quiet little signal, which always hovers close to 0v, it will sound pretty awful. For this reason, pure Class B is rarely used in audio amplifiers.

Class AB is where you take a Class B amp and feed the transistor pair just enough extra electricity to keep them from fully shutting off at 0v. Instead, the positive one will stay on until it gets a little negative, say, -1 volt, and the negative side one will keep going to +1v. This keeps the transistors working a little harder than in Class B, but not by a whole lot, and the crossover distortion becomes less of a problem. In essence, it has become Class A so long as the signal is between +1v and -1v, and Class B the rest of the time. Because it avoids the problems of both Class A (power wasted as heat) and Class B (ugly distortion), it is extremely common in the output stages of power amps, including all larger tube guitar amps, and nearly all solid state guitar amps.

But there's every shade of Class AB. You might only feed it a tiny bit of extra power, so that it's only Class A between +0.1 and -0.1 volts; or you might give it loads, keeping it Class A from +10v to -10v.

In hi-fi audio, I would have to say that Class A is the best, despite being very inefficient. It is just very pure and undistorted sound. With guitars, it's a matter of taste, and not one with a single, clear answer. Most of the amps whose tone you've loved have had Class AB output stages, but a few may have been Class A. Pure and undistorted is a virtue sometimes, but at other times you don't want it at all. And even great sounding clean amps, like most Fenders and Hiwatts, can have AB output stages. Apparently the designer whose promotional materials got us onto this topic thinks that a hotter-running Class AB is better for guitar, but not everyone shares this opinion.

[Andy S]

Ah!!! I had a feeling about the Alpha Omega post..... oh well, regardless... I don't have time right now to reply to my comments as I just got a call for Detroit, Michigan. I'm in Cincinnati, OH and it's a 5 hr drive to the account. I'd like to be back in time to see OSU play LSU in the BCS game.

But P-90 did a great job with the explanation. I'll add later why I commented as I did and maybe a few qoutes from both authors of the books I have. It's more of a marketing issue and the use or misuse of words the authors have a problem with.

Andy S.

[Andy S]

One of the Books I have is "The Guitar Amp Handbook, Understanding Tube amplifiers and getting great sounds" by Dave Hunter.

The other is, "The Tube Amp Book by Aspen Pittman. He is the founder of Groove Tubes.

OK, after reading and rereading "The Guitar amp Handbook" section on Operation classes, the Class A is usually a smaller amp. It is always a single ended amp. The output section is always "on", that is, in AB mode, there is one point where the circuit comes to a point where one section essentially stops working, just for a split second. It makes for a more efficient amp.

It seems it is mostly both authors have an issue with manufacturers stating an amp is Class A even if it is AB but runs in a mode that has both halves operating without shutting down. Thus, a person can probably say it is "Near Class A" operation.

What I find really cool, is that there have been a bunch of small, actual class A amps made recently, for NOT huge amounts of money, that can put out some really good tone.

In my case, anything loud enough to bleed through the floor of the family room from the basement, is too loud and would get the occupants from the family room stomping there feet. Not in time to what I'm playing, but as a signal to cease & desist!! HA!

Well, the big game is starting, I'm out of here!!

Oh yeh, I'd recommend either of the books if you are interested in amps. The Tube Amp book is a lot of history on amps in the first half.

The second half is all technical stuff. Quite detailed.

In the "Handbook", he gets in some details but explains it in laymen terms. So you can get an understanding of what happens. There are also some neat interviews with founders of some of the "Boutique" amp companies and how they started in the biz.

Well, it's time for the kickoff, gotta go!!

Andy S.

[P-90]

Hmmm... I, too, have major qualms about marketing departments calling Class AB amps Class A, but they're very different from those expressed by the authors you mention.

Class A differs from B/AB in bias, but does it really in topology, i.e. can't a push pull topology, whether complementary or quasi-complementary, still be Class A if it's biased high enough? Some excellent designers have thought so, and said as much even when they had no incentive for stretching the truth. Nelson Pass, for example, designed a complementary topology "Class A" power amp which I built (and modded, of course) back in the mid-'80s. Absolutely GORGEOUS sounding amp, although it required about 30 pounds of power supply, over a square foot of 4" deep heatsinking, transistors which had a combined rating of 2,000 watts, and it produced only 20 watts per channel. But the bias was Class A; no transistor ever approached cutoff unless its opposite-polarity partner were equally close to clipping. And it heated the room quite nicely.

Class A differs qualitatively from Class B, or AB, in two major regards:
(1) much less distortion, and little or no need for negative feedback
(2) the distortion that there is, is primarily benign-sounding second harmonic

A push-pull Class A design fits #1, but not #2, since push-pull topologies cause most even-order distortions which occur in the output stage to cancel each other out. Is that important? In a hi-fi amp, which produces less than 0.1% harmonic distortion at full blast, most of it produced by preamp stages which are single-ended Class A, I seriously doubt that I could tell the difference.

(Bridging amplifiers also causes cancellation of even order harmonics. If you bridge two single-ended Class A amps, do they suddenly stop being Class A? If so, what class are they?)

Bias, on the other hand, makes for differences which are easily audible by anyone who is not stone deaf. Otherwise, I would never have considered building what amounted to a 2000 watt amp with 40 watts of output. The usual bias on a 100 watt Class AB amp only keeps it in Class A mode for a fraction of a watt, so we're talking a couple of orders of magnitude difference between that and Class A. Marketing departments do lie about this sort of thing, too -- calling their products "near Class A," or "rich Class AB," etc., when all they've done was to go from 0.1 watts of bias to 0.5 watts. That's deceptive, and pretty pathetic IMO. If they want to sell a Class A amp, they need to either bite the bullet and spend huge bucks on materials, or settle for very low output, there is no other (honest) way to arrive at that goal.

Of course, none of this will matter to people who play electric guitar, since we quickly tire of 0.1% distortion.