also no response to this picture that clearly shows you don't need better amps at low power:
https://www.diyaudio.com/community/...g-compensation-comparison.418073/post-7804955It's funny.
For several days now, some people have been talking about the effect of a class D amplifier on AM radios to prove that they radiate.
But no one talks about the interference of these AM radios, WIFI, cell phone, digital power supplies of computers and others remote controls,on the few milivolts of signal from a vinyl record player.
However, they have a certain influence in radiation but also in conducted interference on the 110 / 230V electrical network.
For several days now, some people have been talking about the effect of a class D amplifier on AM radios to prove that they radiate.
But no one talks about the interference of these AM radios, WIFI, cell phone, digital power supplies of computers and others remote controls,on the few milivolts of signal from a vinyl record player.
However, they have a certain influence in radiation but also in conducted interference on the 110 / 230V electrical network.
With regard to the extreme loudness optimizations (the sound mush that must inevitably result from the destroyed data stream!), we should all ask ourselves whether 2 * 6W (real watts) are not completely sufficient to get on our own nerves with the sound mush.
I don't need a switching amplifier for 6W - Class D high power superduper
Incidentally,
Nick's new throw is intended to compare the classic representatives B, AB, doubleCross and in particular Nick's interpretation (taken view) of the AB-BA concept /interpretation. So we are only interested in one technical matter in this thread:
the design of the last stage of an audio amplifier, a so-called power amplifier, as EF2+n parallel branches. Actually what Bob C. calls a triple.
No more and no less, we can reduce it to the matching of the respective NPN-PNP complementary pair, i.e. the processes at the input characteristic (3rd quadrant of the classical 4th Q. thing) - the optimum bias voltage!
Furthermore, we should explain what a BJT exactly is and how it behaves - otherwise nobody can really understand the wingspread thingy. Honestly, you don't necessarily need this diagram, it's quite nice - but nothing more. But let's put this topic to rest and wait for @Nick Sukov ... alternatively, we could just be happy about another functional audio power amplifier.
greetings,
HBt.
🧙♂️
I don't need a switching amplifier for 6W - Class D high power superduper
Incidentally,
Nick's new throw is intended to compare the classic representatives B, AB, doubleCross and in particular Nick's interpretation (taken view) of the AB-BA concept /interpretation. So we are only interested in one technical matter in this thread:
the design of the last stage of an audio amplifier, a so-called power amplifier, as EF2+n parallel branches. Actually what Bob C. calls a triple.
No more and no less, we can reduce it to the matching of the respective NPN-PNP complementary pair, i.e. the processes at the input characteristic (3rd quadrant of the classical 4th Q. thing) - the optimum bias voltage!
Furthermore, we should explain what a BJT exactly is and how it behaves - otherwise nobody can really understand the wingspread thingy. Honestly, you don't necessarily need this diagram, it's quite nice - but nothing more. But let's put this topic to rest and wait for @Nick Sukov ... alternatively, we could just be happy about another functional audio power amplifier.
greetings,
HBt.
🧙♂️
Last edited:
In the article they are not clear, all they say is that the main residual is at -20dB below the audio 1kHz reference signal.
Since we know that the Hipex is at -35dB and 0.4V we can conclude that in this case the reference is at 56V audio signal,
i would imagine that Purifi s measured -20dB is at about 10V audio signal since the reviewer make measurments at 10W/4R,
this would amount to 1V main residual, wich is comparable to the Hipex s 0.4V.
Beside here what the reviewer mentioned :
Generaly when a manufacturer doesnt provide a number for a key parameter it means that it s not a favourable number,
that s like a car manufacturer that wouldnt publish his cars CO2 rejects.
No need of a vinyl player, here, and for years, you could listen to FM radio France Info just by connecting a 50cm
cinch cable to the selected line level input of an amp, and yet FM is supposed to not be as intrusive as AM, but
they had high audio signal level and there s some AM modulation that occur because they are close to saturation.
At the time it was TV channels that were a nightmare because of France SECAM standard that used AM for the audio,
btw i m at 4km distance of a 190m structure wich handle all official TV/Radio relays.
Please save your thread
Dear @Nick Sukhov !
Every "communications engineer" knows what extremely fast rising edges are - including switching operations.
The whole topic of electrosmog should be discussed elsewhere, apart from that, it's all over anyway.
Since the public debate on the subject of EMC and the omnipresent smog began in the early 1990s, the situation has not changed for the better. It is directly comparable to the climate change predictions of the early 1970s, ... we know it and yet we don't want it to be true.
#
But
our basic topic is actually Nick's AB-BA 150W hifi amplifier with simplified doubleCross-X-Over structure - namely nothing more than a simple push-pull power amplifier.
And
Nick's desire to finally counter the supposed Class-D dominance with quality (that causes no damage).
Do I see that correctly?
HBt.
Dear @Nick Sukhov !
Every "communications engineer" knows what extremely fast rising edges are - including switching operations.
The whole topic of electrosmog should be discussed elsewhere, apart from that, it's all over anyway.
Since the public debate on the subject of EMC and the omnipresent smog began in the early 1990s, the situation has not changed for the better. It is directly comparable to the climate change predictions of the early 1970s, ... we know it and yet we don't want it to be true.
#
But
our basic topic is actually Nick's AB-BA 150W hifi amplifier with simplified doubleCross-X-Over structure - namely nothing more than a simple push-pull power amplifier.
And
Nick's desire to finally counter the supposed Class-D dominance with quality (that causes no damage).
Do I see that correctly?
HBt.
Nick s methology is to reduce the OS distorsion, but at some point the previous stages distorsion will
be the imiting factor as both IPS and VAS GNFB contribution will be limited by these stages own linearity.
OSs distorsion s perceived as high because tests are done within an open loop, but the VAS distorsion
is also high when measured in isolation without NFB.
be the imiting factor as both IPS and VAS GNFB contribution will be limited by these stages own linearity.
OSs distorsion s perceived as high because tests are done within an open loop, but the VAS distorsion
is also high when measured in isolation without NFB.
wahab - That is true in most amplifiers, but the distortion of class A stages can be made arbitrarily small through degeneration and local feedback. Only the final class AB output stage must be non-linear. IMO, spreading out the class AB non-linearity is worthwhile as long as other parameters don't become worse.
Ed
Ed
On page 140 of the “Audio Power Design Handbook”, Figure 6.1, we can see what the distortion spectrum of the EF2 looks like. From underbias to overbias ..!
fifth edition
Douglas Self
What am I getting at practically? That we need to look at the resulting distortion spectrum, or analyze the residual ..!
#
It is the tiresome topic - and we are going round in circles, but we should agree that the isolated consideration (i.e. the path that Douglas S. had taken as an example) of different mechanisms leads to the goal in any case / guaranteed.
Straight ahead and forwards.
From now on, I will relegate everything else to the realm of competence disputes and vanity.
##
We could only discuss Nick's new offer on the DIY market of audio amplifiers if we could find all the details of his construction proposal here, but this will never be a scientific treatise.
###
In the meantime, I worked out the DoubleCross topic myself, it wasn't difficult.
In plain language this means for me
without Nick's guidance and moderation of his/this thread, I don't know what we should be talking about anymore.
The five (?) discussants in this thread are experts themselves and can also create a SuperDuper all by themselves. I'm only interested in the simple beauty of a design, but of course it has to function flawlessly. I think this is exactly what the author Douglas Self associates with the term “blameless”, where it sublimates itself.
I don't like Bob's BC1 design at all - even though it produces first-class results.
That's life.
kind regard,
HBt.
fifth edition
Douglas Self
What am I getting at practically? That we need to look at the resulting distortion spectrum, or analyze the residual ..!
#
It is the tiresome topic - and we are going round in circles, but we should agree that the isolated consideration (i.e. the path that Douglas S. had taken as an example) of different mechanisms leads to the goal in any case / guaranteed.
Straight ahead and forwards.
From now on, I will relegate everything else to the realm of competence disputes and vanity.
##
We could only discuss Nick's new offer on the DIY market of audio amplifiers if we could find all the details of his construction proposal here, but this will never be a scientific treatise.
###
In the meantime, I worked out the DoubleCross topic myself, it wasn't difficult.
In plain language this means for me
without Nick's guidance and moderation of his/this thread, I don't know what we should be talking about anymore.
The five (?) discussants in this thread are experts themselves and can also create a SuperDuper all by themselves. I'm only interested in the simple beauty of a design, but of course it has to function flawlessly. I think this is exactly what the author Douglas Self associates with the term “blameless”, where it sublimates itself.
I don't like Bob's BC1 design at all - even though it produces first-class results.
That's life.
kind regard,
HBt.
Is there any consensus as to what output power has to be necessary before class D is honestly dominating the SOTA with an analog input and when efficiency simply doesn't matter? As power falls both types linear and switching get easier but at some point doesn't class D just become extra electronics with more problems to handle?
Degeneration work up to a given amount, improving the linearity of the class A stages by mean of local NFB will reduce
the available GNFB used to linearize the OS, beside a non degenerated IPS FI will still be as linear because its higher gain
will still linearize it through GNFB as much, if not better, than local degeneration all while better linearizing the OS.
In this respect TMC compensation is of big help since it localy linearize both the VAS and the OS, only limitation
is that we cant use more than 10dB local NFB enclosing the OS without endangering stability, as for the IPS you
surely noticed that about all op amps input differentials have no degeneration, and that s for a reason.
The primary reason of omitting degeneration resistors inside op-amp input stages is avoiding additional input noise contribution imho.
This is a legitimate question, which I would like to answer with an analogy or a comparison:
Two master chefs prepare a very specific and well-known meal, both use the usual ingredients - in the end, both dishes are excellent and yet Chef A, with all due respect to Chef B, prefers his own cooking times and small trademarks - including the quantities of seasoning used. Overall, he likes his beauty a little better.
I would like my statement to be understood in this sense.
HBt.
"imho" (typisch deutsch)
In other words, you don't know!
I wouldn't be surprised if "Bernhard" turns up soon ;-).
greetings,
HBt.
So what - should I pretend things I am not sure about? And btw, I am sure that input degeneration shows this effect - but I am not in the position of a chip designer. Furthermore, I am a typical German. I am not specially proud of that, but no need to be ashamed either. Your attitude smells a bit snobby to me , uhh? And that's it.
Last edited:
The inflationary use of stupid abbreviations has sadly become a German trademark.
That's all.
I'm just noting that you've gone off into the blue with the noise.
Our current topic touches on the linearization possibilities of the non-linear input characteristic of a BJT. That's what it's all about at the moment and the closed loop..!
That's all.
I'm just noting that you've gone off into the blue with the noise.
Our current topic touches on the linearization possibilities of the non-linear input characteristic of a BJT. That's what it's all about at the moment and the closed loop..!
Is that a fact ? Hmmm ... wonder whether that carries out to larger amplifiers ? I have 2 OEM designs that run at very low current in the LTP , with
no degeneration resistance - ABSOLUTE silence. I wondered what design choice contributed to this.
Quieter than (all) of my DIY PPM amps.
OS
I consider a typical matter to be reprehensible, the subsequent modification of one's own contribution as a subsequent reaction to a critical follow-up contribution. That is a falsification - and by the way, we can also settle our personal disputes behind the scenes.
feel free to send me a PM.
HBt.