What ever happened to Class-B?

[JoeBob]

Well, I finished off my first two poweramp projects (IC based solutions) and now that I've learned alot and my discrete preamp's comming along, I thought my next undertaking would be a nice discrete BJT poweramp.

I read alot about the different designs, classes and all. And while Class-A obviously has no crossover distortion, I just don't understand why it's so popular, it's just way too inefficient (but, then again, that's just how I see things). And there are obviously all those numerous Class-AB designs around. But it seems to me a correctly done pure Class-B amp would be better then a Class-AB amp. And the idea of Class-B apeals to me, if it could be done properly, don't get me wrong, I like good sound quality.

So it leads me to ask, anyone know of any good resources for information on pure Class-B poweramps? And maybe some quality pure Class-B designs I could take a gander at?

I just don't see why all the discussion is about Class-A amps. Doesn't anyone find the idea of turning all the power into heat a waste but me?

 

[mrfeedback]

Hello JoeBob, Doug Self did some exploration of class B with good results from memory, and yes the idea of two 300W room heaters that give 0-100w of audio output goes totally against my tree hugging hippy evironmental and conservational ideals.
I read a short article a few years back stating that the US has had to build something like 40 extra power generating stations to supply the power demanded by PC's in offices and homes and kept running 24/7.
I suppose if I owned a power company I'd happily promote class A ! (and PC's 24/7).
And besides the addition to the household monthly power bill is NOT insignificant.
Maybe in cold countries Pass etc amps allow you to turn off the central heating, but that does not apply here in Australia !.

Peace Brother,
Eric the 40YO hippy.

 

[GRollins]

I once heard someone say (on the subject of pants suits for women) that they were okay...for wearing to a dog fight. That's how I feel about class B.
No one contests that class A is inefficient. It's simply a question of how many corners you want to cut. Class A costs more, primarily in the power supply and heat sinks but, all factors being equal, sounds better.
Apologists for class B frequently claim that it sounds 'as good as' class A. Why do they feel the need to compare to class A? If class B is so good, why aren't people saying that class A amps sound 'as good as' class B? With enough feedback you can get specs that have any arbitrary number of zeros after the decimal point that you might wish. They still don't sound as good.
It comes down to two things: money, and how critical a listener you are. The money thing I can sympathize with--if I had to start an amp right now, it'd probably be class C...maybe class Z. The critical listener thing trips up a lot of people. For reasons of peer pressure, pride, or whatever, a lot of folks say they want class A, but they either can't hear the difference, or the rest of their system isn't good enough to let them hear it. You are allowed to say that you don't care about that N'th degree of performance. Honesty is good.
The choice is yours. It's your money and your ears.

Grey

 

[JoeBob]

Ya well, the increase in performance I feel isn't worth the extra cost/heat/weight/hassle. But I still would like something good.

I haven't read much on Class-C and haven't even heard of Class-Z. I guess I should find some good readings on those two. I just thought there'd be more info on quality Class-B designs, since well, Class-B has been around a while, I thought it would have matured more then those funky new Class-D, Class-G, Class-H, Class-T and what not designs.

 

[jam]

The best resource for class B amplifiers is Doug Self's book and his series of articles in Electronics World.
I am in total agreement with Grey, and to me the best use for class B amplifiers are motor drivers (hard drives and servo systems) and have little use in High End audio applications, Even cheap Japanese recievers use class AB designs. Getting a class B amplifier to perform correctly requires too much complexity.

Jam

 

[GRollins]

(class C is equivalent to 100% clipping, all the time--good only for radio transmitters...I made class Z up on the spur of the moment trying to be funny, there is no such thing as class Z)

Grey

 

[JoeBob]

Thgouht so at first, but wasn't too sure .

 

[leroy]

Why would anyone build a 'pure' class B amplifier? Everyone knows class A/B is the better solution when it comes to efficiency and output power.

Joebob, let me give you advice: get rid of the 'pure' class B idea and start thinking about a digital amplifier, then you're really heading for high efficiency.

leroy

 

[promitheus]

I have made a lot of class B amps specially from Elektor magazine. If they are a good design like from Elektor they sound very good. Elektor uses a double differntial input which makes the wholoe amp symmetrical till the load. Class A sounds a lot better, but you need a good system to hear the difference. Not a very expensive one a good one. Clear preamp not complicated and some good speakers.
What I find very attractive are digital amps. They are very simple in design, almost like the Pass amps, they accept directly digital input from a CD and they can be directly coupled to a speaker load without an extra crossover.

 

[vbd]

Hi all,
Actually, all designs reffered to as class B in this thread are class AB, where there is a small overlap in the conduction of the push pull transistors, to avoid the horrible crossover distorsion...
The point of determining if class A or class AB is better is a matter of personnal choice/budget/and so on... IMHO, a well designed and constructed class AB amp is OK, but PURE class B is not a good thing...
The only good amplifier I know with a pure class B output stage is the now classic QUAD 405, where the crossover distorsion is cancelled (or quite...) with a quite complex feedback/feedforward circuit arrangement, but maybe some of you know other ones...

 

[mefinnis]

G'Day ...... as we say

I have built a number of Leach amps and these are (IMHO) quite respectable Class-AB units. You would likely be very happy with one and there would be a lot of help available on the web: http://users.ece.gatech.edu/~mleach/lowtim/

Many have built and recommend the ASKA amp. Haven't heard one myself but it certainly bears consideration: http://www.printedelectronics.com/pe/gomain.htm

Also, from a recent thread there is a forum for these units: http://www.harmonicdiscord.com/forums/viewforum.php?forum=42&828

Rod Elliot also has a committed following and may be worth a look: http://sound.westhost.com/projects.htm
He has a number of amplifiers, including the much loved "Death of Zen".

WRT Class A - I have to agree I prefer this, but the costs are significant, and in my earlier years (where cost, space, heat, etc mattered a lot) I was quite happy with the WML amp.

Cheers and have fun !

Mark

 

[Bill F.]

New topology Class B

I posted this a while ago and got no takers. Maybe I just posted it in the wrong forum. Any takers now?

http://www.angelfire.com/ab3/mjramp/

>Can anybody comment on this class-B topology by Mike J Renardson?

>I've been thinking about a loudspeaker driver with dual tandem voice coils each connected to a class-B output. Each coil would handle 1/2 of the waveform while the other, non-energized coil leaves the magnetic gap. The net effect should double the linear excursion of any magnetic motor while maintaining efficiency. Proper positioning of the coils should dial out crossover distortion.


Bill

 

[FEThead]

pure class B ?

Hi JoeBob

I fooled around trying to build an amp that was pure class B (or as close as I could get) a few years ago. One thing I "discovered" is that active devices (vacuum or silicon) are pretty non-linear around cut off. At low levels in "pure" class B the distortion is pretty bad. Class B looks good on paper but does not go into practice very well. That is why people run class AB.

Happy Listening

 

[vbd]

Re: New topology Class B

>http://www.angelfire.com/ab3/mjramp/

>Can anybody comment on this class-B topology by >Mike J Renardson?
Hi,
It makes me think of the old 741 op-amp trick, with a resistor from output to minus supply, that gives a similar result as Fig 4 of http://www.angelfire.com/ab3/mjramp/pagefour.html
The top transistor has an Iq, and the bottom one has no Iq... But here, where does the Iq go ? In the load, to create a DC offset voltage ??? I must try this curcuit !!!....

 

[pmkap]

So I guess none you young'ns are old enough to remember the Blomley amp design? (Wireless World Feb, Mar, 1971) Driving the nasty impedance of that common emmiter guiotine was difficult back in the day, but with Zero cross detectors and a fet buffer might be do-able today (but certainly beyond my ken). As it still maintained a fixed small bias even on the inactive side, it might not be considered class B. A precusor to Self's blameless [mislableled] class B, with that suckout cap?

 

[subwo1]

>http://www.angelfire.com/ab3/mjramp/

>Can anybody comment on this class-B topology by >Mike J Renardson?

Someone alluded to what my impression also is. The output transistor on-bias is basically carried to a later portion of the circuit.

 

[JoeBob]

>http://www.angelfire.com/ab3/mjramp/

>Can anybody comment on this class-B topology by >Mike J Renardson?

Well, he originally wrote an article about this in Electronics World magazine. I ordered the archive and plan on reading through it. There's also another interesting Class-B article in the archive, some other member of this forum (sorry, can't remember who) thought quite highly of it. I plan on reading both (as soon as they get here).