Class B amplifier with SINAD of 120, how is that possible?

[restorer-john]

In case you guys are interested, I will prepare a listening test with various level of crossover distortion, based on output stage bias of the old class AB amplifier topology. Measurements will go together with music samples.

Probably not necessary Pavel. People don' care for endless ABX tests and presenting their results for scrutiny. Engagement in such "tests" is understandably poor.

 

[Mnyb]

Eh have anyone actually confirmed that it really is class B ? If it has bias current at all will it not by definition be class-AB

The premise of this whole tread could be nullified

 

[restorer-john]

If it has bias current at all will it not by definition be class-AB

I don't really think anyone seriously cares one way or the other. I'm somewhat curious, but not curious enough to plunk down a pile of my own dollars to investigate WTF Topping is doing with yet another toy amplifier in a tin can powered by a plugpack. It's chicken-sh#t stuff in the scheme of things.

 

[SIY]

If it has bias current at all will it not by definition be class-AB

I think you mean the opposite.

 

[Blumlein 88]

What compels you to repeatedly post condescending crap like that?

Well apparently you don't know everything. Apparently amplifier circuit designers know everything (about amplifier circuits). Including every possible circuit design ever designed otherwise how could they design without knowing design. Somehow apparently all their minds are quantum entangled and for one anywhere to know they all know. So if you don't know you aren't one?

 

[Mnyb]

I don't really think anyone seriously cares one way or the other. I'm somewhat curious, but not curious enough to plunk down a pile of my own dollars to investigate WTF Topping is doing with yet another toy amplifier in a tin can powered by a plugpack. It's chicken-sh#t stuff in the scheme of things.

Some do like the OP but I think some people are to worked up about amplifier classes ? Do they sort thier underwear by color .
To me as a non amplifier designer I care more about the end result, but are slightly curious on how the sausage is made .

It could be class-Z to me . But it’s part of marketing and hifi lore , there are many high biased class ab amps touted as class a . And marketing in the past have not made things easier like class AA or new class A or whatever fancy name you give a bias and feedback scheme ?
Duo beta circuit? Was that luxmans marketing name for a multi pole frequency dependent feedback loop ?

 

[Mnyb]

I think you mean the opposite.

My bad engrish

 

[Sokel]

Well apparently you don't know everything. Apparently amplifier circuit designers know everything (about amplifier circuits). Including every possible circuit design ever designed otherwise how could they design without knowing design. Somehow apparently all their minds are quantum entangled and for one anywhere to know they all know. So if you don't know you aren't one?

When I built my house an architect,a civil engineer and a structural engineer involved.
Each one cursed the other two all the time,I will be lucky if that thing won't become my grave.

You should have known by now

 

[SIY]

Well apparently you don't know everything. Apparently amplifier circuit designers know everything (about amplifier circuits). Including every possible circuit design ever designed otherwise how could they design without knowing design. Somehow apparently all their minds are quantum entangled and for one anywhere to know they all know. So if you don't know you aren't one?

And worse yet, I design amps so I’m automatically a suspicious character.

 

[Old_School_Brad]

I'll just stare at the ground and drool.

What compels you to repeatedly post condescending crap like that?

I don't mind. I, for one, am not an amplifier circuit designer. And I occasionally drool. Mostly in my sleep.

 

[SSS]

I don't really think anyone seriously cares one way or the other. I'm somewhat curious, but not curious enough to plunk down a pile of my own dollars to investigate WTF Topping is doing with yet another toy amplifier in a tin can powered by a plugpack. It's chicken-sh#t stuff in the scheme of things.

Right, too much Topping toy amplifiers here. By the way, as far a I learned it, class-B is when no quiet current flows. Class-C is when no quiet current flows and for instance a sinewave is cut around the zero value. Class-D amplifiers are mostly used in radio transmitters.

 

[pma]

In case you guys are interested, I will prepare a listening test with various level of crossover distortion, based on output stage bias of the old class AB amplifier topology. Measurements will go together with music samples.

OK, I am working on it. It is a class AB topology from 1969, slightly improved and with better parts. It has 0R22 Re resistors. The worst scenario is with 0mV bias voltage across R and it looks like this, at 1.5W/4R7 load. This will be compared with class AB optimally biased, which is at 25mV across Re, about 100mA idle current. A usually and as is my habit, I am not messing up the signals with ugly 50Hz mains and its multiples, that can be seen in almost all older linear amplifiers here.

 

[rongon]

I replied to a reply in the original review thread, but I see this one's been created, so I'll post it here:

T'ang-ku-la Module
The creation of T'ang-ku-la module shows our determination to pursue excellence and constantly challenge the physical limits of performance.

The T'ang-ku-la Module uses a brand new three-stage feedback circuit technology. It reduces circuit noise and high-frequency distortion, thereby increasing the dynamic range of the power amplifier to 151 dB. This technology also significantly lowers the heat produced by circuits, further boosting the reliability of the circuits, extending service life, and effectively diminishing thermal noise.

That's marketing speak. It doesn't clarify whether the amp is biased in class B or class AB operation. It just makes a vague claim about using three stages of feedback (Positive? Negative?) without further details. Are those feedback loops nested? Or local to each of three stages? Or three global feedback loops? (That would be strange.) You can't tell from the marketing blurb.

Topping says is class B.
Some sellers advertise it as class AB.
There's no mention in the spec's sheet.

Where does Topping state that this amp's output stage is biased to class B operation? It's not in the spec sheet? That's a pretty basic specification. This is all kinda strange.

Compare a good class B based amplifier that is designed with what you mention: Using better reacting output devices, smart biassing of the output devices (not switching completely off), feedback trickery and lots of open loop gain and high BW could all be part of the solution.
....against a good class AB based amplifier. In such a case, what does a class AB based amplifier add that a class B does not have?

I could be wrong (maybe SY can clarify here) but I thought class B biasing of a push-pull output stage is defined by the output devices being biased to cutoff at zero input, so quiescent current draw is near zero with no signal applied.

If you use output devices that don't switch completely off (i.e. they conduct current even with no signal applied) wouldn't that mean they're biased to class AB, and not class B?

I was just reading about it in the old Radiotron Designer's Handbook 4th Edition last night. I'm also trying to get my head around the idea of a class B amplifier that doesn't exhibit crossover notch distortion.

 

[pma]

Probably not necessary Pavel. People don' care for endless ABX tests and presenting their results for scrutiny. Engagement in such "tests" is understandably poor.

Yeah, you are right. Anyway, I was curious. The test files, together with delta (difference) file are here:

xoverdist.zip

drive.google.com

"Optimalbias" is with 22mV across 0R22 Re resistor (100mA current), "zerobias" measures 0.0 mV across Re resistor, so there may be some negligible current as the lowest measurable voltage was 0.1mV (0.45mA current). Spectra at 1.5W comparison:



Though the deltawave delta file sounds ugly (but only after huge unrealistic amplification), the files are indistinguishable to me by listening.

Paul's @pkane excellent Deltawave software says:



The crossover distortion should have been much higher to be audible. Even in this 55 years topology a moderate feedback factor made the crossover distortion inaudible with music.

 

[wwenze]

Class D is like crossover distortion to the max? With feedback it behaves like "oh i need to raise the output node voltage BRGHUAGHAGH ok I've done enough imma go sleep now"

My train of thought may be weird but I'm always thinking about the possibility of switched linear-region or PAM output stages for either performance or efficiency improvements.

 

[DanielT]

I don't really think anyone seriously cares one way or the other. I'm somewhat curious, but not curious enough to plunk down a pile of my own dollars to investigate WTF Topping is doing with yet another toy amplifier in a tin can powered by a plugpack. It's chicken-sh#t stuff in the scheme of things.

Some do like the OP but I think some people are to worked up about amplifier classes ? Do they sort thier underwear by color .
To me as a non amplifier designer I care more about the end result, but are slightly curious on how the sausage is made .

It could be class-Z to me . But it’s part of marketing and hifi lore , there are many high biased class ab amps touted as class a . And marketing in the past have not made things easier like class AA or new class A or whatever fancy name you give a bias and feedback scheme ?
Duo beta circuit? Was that luxmans marketing name for a multi pole frequency dependent feedback loop ?

I'm just a little curious how a class B amplifier can be designed to get such good performance.

In practice, I don't choose amplifiers based on amplifier class. Other factors then matter, such as: Price-performance, power level, good FR (not load dependent), sensible SINAD (adequate ditto), functions and durability. Nothing revolutionary. On the contrary, I think they are fairly common standard criteria/factors that many weigh together in their choice of an amplifier.
Physical size and good looks can be thrown into play, even if those factors are of secondary importance to me.

 

[solderdude]

You can switch them off however you must not use Re additional resistors. This is a fact and amplifiers like this were built.

NAD built those and they were plagued with thermal runaway.

 

[solderdude]

.against a good class AB based amplifier. In such a case, what does a class AB based amplifier add that a class B does not have?

AB amp will be simpler to build but will run hotter in idle.

 

[DVDdoug]

Class D is like crossover distortion to the max?

No. It's more like a switching power supply where the MOSFET switches on & off at a high frequency (trying to never be partially-on) but the output is filtered to pure-constant DC.

Yes, in both cases corrective feedback is involved

 

[pma]

NAD built those and they were plagued with thermal runaway.

They did not do it properly. It can be done much better, without the risk of runaway.