Class D = Trash?

This is a statement from Optoma USA customer help, a world wide company who I believe handled NuForce, maybe still do.

Digital Switching Amplifiers (commonly known as Class-D) have been around for years. Nevertheless, it is nearly impossible to engineer a conventional Class-D amplifier that handles the full requirement, 20-20,000Hz, for full-bandwidth music reproduction. A Class-D amplifier works by utilising a high-frequency saw-tooth waveform to modulate the music signal (to learn more about how Class-D amplifier works, click here).

The constant presence of the saw-tooth waveform, which is very high in frequency spectrum and its inevitable frequency jittering, can mask or corrupt low-level music signal. The output filter designed to filter out noise and overtones caused by the saw-tooth waveform adds a 180 degree phase shift to Class-D output stage, causing possible instability and adding distortion due to its own inherent non-linearities.

Additionally, the output filter presents frequency-variant output impedance that can interact with a speaker’s complex impedance. Variants of Class-D amplifiers with the addition of Digital Signal Processor claim to improve music reproductions.

However, because of their lack of close-loop design, especially from the speaker’s terminals, spurious interaction between the speaker’s complex impedance and back-EMF with the amplifier’s resonant output filter can result in harsh sound reproduction.

The fundamental flaws of conventional Class-D amplifiers remain unresolved.

Cheers George

atmasphere,

      Thank you for replying to my questions.

    Your confirmation that current class D amp switching frequencies are not causing 'inharmonic distortion', along with class D Ice and Hypex NCore power module inventor Bruno Putzeys' statement that current class D switching frequencies in the mid-500kHz range are "completely reasonable", refutes the theory that they need to be raised to the 3-5mHz range (to not affect frequencies in the audible range) as far as I'm concerned.
     These confirmations make sense since they mirror the more subjective perceptions many class D users have, including myself, of the total lack of sonic anomalies and very high quality performance levels of class D amps..

    I'm looking forward to hearing the results of your first class D amp design effort.
   I suspect your class D amp will be transparent, detailed, with a very neutral overall quality while also having the extreme low noise and distortion levels that all seem to be common characteristics of good class D amp designs.  
   I believe you'll also find that an amp with the above traits will be a great match with high quality tube preamps since the amp will operate as the audio ideal of 'a straight wire with gain' which  will allow the sound qualities of your tube preamps to be amplified faithfully without any alterations. 

Tim
    

no-bell - instead of wasting time on your verbose post above, you could simply use google to search for "inharmonic distortion"

try it

(and, no, I don't claim that Class D is good or bad -- I DO want to know if it offers any advantages for me)

1. Is there a consensus among amp designers that current switching frequencies (what I believe you're calling 'scan frequencies') being too low is the primary cause of 'inharmonic distortion'? If so, is there also a consensus that raising switching frequencies to the 3-5 MHz range would reduce 'inharmonic distortion'?

No.

2. Can you explain why I, and apparently many other class D amp users, do not perceive our class D amps as being cold or sterile? Do you think the ability to perceive the affects of 'inharmonic distortion' differs among individuals or do you think it's more likely the degree of 'inharmonic distortion' varies by amp?

A lot depends on what your reference is and different amps have differing amounts of distortion. Like anything else in this world, you have to try it. As a general rule of thumb though, if you can turn it up loud and it also **sounds** loud, then you know you have a problem. If your system is devoid of higher ordered harmonics and inharmonic artifacts, you won't know how loud the system is until you find that you have to yell at someone sitting right beside you to be heard. 

3. Are there any current methods of measuring 'inharmonic distortion'?

Of course! The best use is with spectrum analysis.

Hello all,

I only time I can recall listening to "Unharmonic distortions" is when our granddaughter is over and wants to play some of her music,she's 12.

Kenny.

randy-11,
     You stated:
"of course, you can measure all sorts of distortion products - inharmonic means non-harmonic."
    Well golly gee, Randy, I would have never thought that 'inharmonic' means the same as 'non-harmonic'.  Of course they mean the same thing, but that doesn't mean that either one is an accepted scientific specification and it definitely doesn't mean either one can be currently measured.
    Despite your claim of certainty, I am going to wait until atmosphere replies before I believe that there's a current specification for 'inharmonic distortion' and whether it's measurable.  I know there's a specification for 'harmonic distortion'  and 'intermodulation distortion' as well as standardized methods of measuring both. 
    But this is the first I've heard of 'inharmonic distortion' and I doubt it's even currently a valid scientific term and, even if it is,  whether there's a standardized method of measuring it.  From atmasphere's earlier post, it seemed to me that the term 'inharmonic distortion' was an informal term used by some amp designers to distinguish it from intermodulation distortion and he made no mention of whether it could be measured.

Erik_ squires,

     I'm in complete agreement with you and I hear absolutely no sonic artifacts/anomalies when listening to my class D amps, either.  I know from personal experience that all of my class D amps outperform my former class A/B amps by a wide margin in every area that I care about, although my A/B amps may not have been as highly biased as yours.
     However, I try to keep an open mind about things and listen to the thoughts and opinions of others, especially from well respected sources such as atmasphere.  Hopefully, he'll impart some more wisdom on this subject soon.

     Ultimately, I realize I'm not concerned at all about whether the theory that current switching frequencies are too low and cause audible artifacts/anomalies is true or not, other than curiosity.  If the theory is proven not true, it will just confirm what I have not been hearing since I began using class D amps in my system.  In the unlikely case that the theory is proven true, I will be admittedly shocked for a few moments but then I'll just continue on enjoying my class D amps as if nothing had changed because nothing will have changed; I will continue to not perceive these sonic artifacts/anomalies that are proven to exist just as I did not perceive these sonic artifacts/anomalies before I even heard of the theory of their existence.      
     Sure, I'll listen to a class D amps in my system that claim, either by employing better passive filtering or higher switching frequencies, to eliminate all  theorized sonic artifacts/anomalies if they ever become affordable just to test if I perceive an improvement in performance. 
    I'm just having a hard time conceiving of how the absence of something (sonic artifacts/anomalies), that I didn't perceive in the first place, is going to improve my system performance.  
     I'm not planning on selling my excellent performing class D amps anytime soon, perhaps ever.

Tim

You know, there are a couple of interesting hybrid Class D which are reminiscent of the carver switching rails designs.

The Yamaha EEEngine and NuPrime. In these cases the voltage rails themselves are Class D which then drive a low wattage linear amplifier sandwiched between them.

Fun times,

E

no worries Eric

I am going to consider an amp replacement after a DAC/disc player upgrade and anything will have to prove itself vs. my current Sunfire (which I guess is closest to a Class G or H)

I am running two Class A/D Auralic Merak mono blocks and also switched from a Parasound A23.  I'm happy, but have no reasonable way to do an A/B test between the Merak's or A23 so I don't know what differences I'm hearing!  I'm getting old!

@noble100  and @randy-11

I suggest you ignore some of these criticisms before listening. There are like a billion claims, many old or outdated or downright false.

I can tell you I cannot, blindfolded, or sighted distinguish between my current amps based on Class D and my previous, Parasound Halo A23's. The latter are linear, hi bias AB. I've also heard megabuck class A amps and mine sound as good (within their volume / power envelope).

For this reason, I completely discredit anyone who claims Class D must have x or y artifact.

Best,


E

of course, you can measure all sorts of distortion products - inharmonic means non-harmonic

Hi atmasphere,
     Thank you for your explanation.  I know your credentials and trust your comments have no hidden agenda.  Unfortunately, I don't know georgehifi's credentials and whether he has a hidden agenda against class D.
     Okay, so you're stating that there can be intermodulations between the scan frequency and the signal being amplified in class D amps.  Amp designers call this 'inharmonic distortion' and its affects may be a cold and sterile sound.  I just have a couple questions about this:

1.  Is there a consensus among amp designers that current switching frequencies (what I believe you're calling 'scan frequencies') being too low is the primary cause of 'inharmonic distortion'? If so, is there also a consensus that raising switching frequencies to the 3-5 MHz range would reduce 'inharmonic distortion'?

2.  Can you explain why I, and apparently many other class D amp users, do not perceive our class D amps as being cold or sterile?  Do you think the ability to perceive  the affects of 'inharmonic distortion' differs among individuals or do you think it's more likely the degree of 'inharmonic distortion' varies by amp?

3.  Are there any current methods of measuring 'inharmonic distortion'?

Thanks.
  Tim

^^ one of the artifacts of Class D is is that there can be intermodulations between the scan frequency and the the signal being amplified.

You can see this in the specs if the spectrum of distortions is graphed. So there isn't a lack of evidence!

In the digital world, they call this 'aliasing'. In the analog world (and since Class D is an analog process) we call it 'inharmonic distortion' to distinguish it from 'intermodulation distortion'; the latter being intermodulations between tones being amplified.

This is why the artifact of Class D is different from traditional solid state. The artifacts are similar to those of digital and you see many of the same complaints leveled against it- cold, sterile, that sort of thing.

Its a simple technology, and that seems in a way to belie the issues of how to correct the artifacts, which is rather tricky. Like digital though it is a rising star in high end audio (many think its already arrived but the fact that this thread exists is evidence that it has a ways to go, otherwise it would have completely supplanted tubes and traditional transistors; since it hasn't, we know without needing to know technical issues that its got a ways to go) and its potential can't be ignored (which is why we've been investigating and working on class D ourselves).

georgehifi,

     Let me get this straight:

     When  I initially asked you:

"1. Does any scientific evidence exist that in any way supports your theory that the current class D switching frequencies are too low and result in sonic anomalies that are audible?

2. Have you personally ever heard these sonic anomalies listening to good class D? If so, please explain what these theoretical sonic anomalies sound like. Apparently, you currently may be the lone known human believing in, and claiming the audibility of, these sonic gremlins."

  Your response, is to post quotes from a very forgettable 10yr old roundtable discussion that actually is more concerned on the relative merits of class D amps than on the existence of audible affects caused by the switching frequencies utilized?  And the panel consists of 9 traditional linear amp (both tube and solid state) designers and a single designer actually designing his amps around class D amp power modules that he did not design but purchased from another company?

 Come On, Man!  You didn't come anywhere near answering either one of my clearly straight-forward questions  by quoting from the comments of a stacked deck panel of traditional linear amp designers, 90% of whom have absolutely no experience designing a single class D amp  

    Who's the genius who picked out this ridiculously unbalanced panel of participants?  Is he the same one who selected the participants for that infamous 'A Remarkably Unbiased Roundtable Discussion on Christianity' that consisted of those 9 adamant atheists and that single choir boy?
     I would have found that Class D Roundtable a lot more informative, interesting and worthwhile had the genius invited an actual class D amp designer, such as Brruno Putzeys, the inventor of both the Ice and Hypex NCore class D amplifier technologies.
     Due to the complete absence of ANY evidence that even vaguely supports your theory, the complete lack of ANY description of what these theorized sonic anomalies  actually sound like,  you're reluctance to state whether or not you've ever actually heard them yourself and my personal experience of never hearing even a hint of sonic anomalies from my class D amps,  I think you understand the reasons you need to come up with a new fake theory/boogieman to further your mysterious agenda.
     How about claiming they spontaneously combust?

Later,
 Tim 

wasn’t that roundtable a decade ago?

I’d listen to modern Class D amps - like the ones listed above. I’d compare using A/B and extended sessions, all double-blinded.

I’d also listen to the Benchmark amp (not Class D), then buy the best one in my price range. if looked ugly, I’d stick it in a maple cabinet and glue some tubes on top of it.

But they will have to pry my tubed pre-amp out of my cold, dead fingers.





Just like the National Righteous Amplification club sez...

Yes, and they are the industries top designers, nothing will change until that dreaded switching frequency is taken up much higher. So then the output filter can do it’s job properly without any effects down into the audio band.

Technics is leading the way with the SE-R1 with twice as high switching frequency, but it’s not mainstream yet, and really should be even higher, one can only hope.

Cheers George

WOW, savage!

Make of it what you will, something more for the OP to read, on the title of his thread.

"Switching Frequency" it’s mentioned a few times here.

This was an Absolute Sounds Round Table discussion with the industry top dogs on Class-D, the only one in favor is the one that manufactures them Jeff Rowland but had little to say amongst his peers.

Bob Carver
"I built many of them right here in my own laboratory with the thought they could and would fulfill that final promise.... I was never able to build a Class D amplifier that sounded as good as a linear one."

John Curl (Parasound, CTC, Vendetta Research, Constellation)
"Some version of hybrid Class A/D looks like the future in optimum audio design."

Cyrill Hammer (Souloution)
"if you want to have your product performing at the cutting edge it is not possible with today’s known switching technologies. In order to come close to the performance of the best linear design we would need high-current semiconductors that provide switching frequencies of several MHz or even GHz."

Lew Johnson (Conrad Johnson)
"I tend to think that Class D circuit design is an approach best relegated to producing low-cost, physically manageable multichannel amplifiers—where one might accept some compromise in sound quality for the sake of squeezing five, six, or seven 100 watt channels into one moderate-sized package for a budget home-theater installation."

Vladimir Shushurin (Lamm)
"No, it is not. And I would like to respond to the second part of this question with an allegory. Any field of human activity defines a number of requirements which, when properly implemented, guarantee a positive outcome.
For example, the basic requirement in the army and sports is an able-bodied individual. So, it would be quite natural to concentrate on searching for such an individual (especially as we know where to find him).
However, out of the blue we decide to choose a feeble-bodied person who, on top of that, is encumbered by various diseases. Having made this decision (which is a priori improper) we start justifying it to ourselves and others by citing the great state of our medicine, which is capable of curing many ailments."

Fumio Ohashi (BAlabo)
"No. Class D can’t really be considered for super-high-end performance in its present stage of development, although it can be fine for mid-market products."

Nelson Pass (Threshold, Passlabs)
"Does a $10 bottle of wine compete with a $100 bottle? Of course it does, and it often wins based on price. Right at the moment Class D designers seem to be still focusing on the objectively measured performance of their amplifiers. I expect that at some point the economics of the marketplace will encourage them to pay more attention to the subjective qualities, and then they will probably play a greater role in the high end."

Jürgen Reiss (MBL)
"I have worked a lot lately with Class D. Ninety-nine percent of Class D circuits are not competitive with linear circuits.
Most Class D sounds sterile. It’s tricky to figure out what to do to compensate for that."

Jeff Rowland
"I consider Class D to be highly competitive in the present, and to offer an evolutionary pathway of audio design that may produce even more astonishing results in the future."

Thorsten Loesch

I have yet to hear a pure class D Amp I’d rate above "below average for solid state" (which is not very high performance).

In a little update of my classic "Valve Analogue Stages for DAC’s #" I wrote:
"Perhaps more crucially, so called Class D Amplifiers, which have in recent times sprouted up like mushrooms after a warm rain, continue to use the straight two or three level modulation scheme described above. And thus they still require the use of heavy handed noise shaping to attain anything like acceptable 16 Bit Audio performance.
The clock frequencies for these amplifiers are usually at 300 KHz to 1MHz in the best cases. That is 3,000 to 10,000 times lower than what is required to attain 16 Bit / 44.1 KHz performance without noise shaping and other forms of signal manipulation!
And again, one is baffled and perplexed by the rave reviews many Class D amplifiers receive, as baffled as one was about the late 90’s reviews of timeslicing dac’s. The best of breed I have auditioned were certainly not bad; however in direct comparison to the best available valve and solid state amplifiers they do not produce a very good sound. Well, at least they offer novelty and the reviewers something to write about other than another (however good sounding) 8 Watt valve amp.
Incidentally, the best sounding Class D amps tend to be really low power single chip devices (putting out little more than the 8 watt valve amps), presumably because they are faster AND because they always work near what one might call “full scaleâ€, if they would be dac’s. On second thought, they of COURSE are DA Converters and where a Class D amplifier accepts analogue input directly it is an A2D converter followed by a power D2A converter!
What an insight!?" Mark Levinsons Interleaving of multiple Class D Amplifiers is potentially a step in the right direction, but does not go far enough.Personally I think that the best option would be something that combines a Class D Amplifier for the heavy lifting with something Class A for fine detail. Probably implemented in the style I did for AMR’s AM-77 "Jikoda$" Style. In this case both of the circuits involved can operate fully open loop.
In many ways the problems in Class D Amplifiers are analogous (but not identical to) those in Class B Amplifiers (but without an option to implement Class AB or Class A) so similar solutions apply.
All Class D amplifiers are essentially delta-sigma DAC’s.
If the input is not digital PWM signals (aka "DSD") but analogue audio then it is also a Delta Sigma Analogue to digital converter...
Now DSD (aka SACD) which to my ears fails to come close, never mind equal true PCM CD Replay in most aspects of sound quality, operates at 2.8MHz switching, or around 10 times as fast as common Class D Amplifiers...
Why anyone would want to listen through an A2D followed by an D2A Converter that are around 10 times worse than single speed DSD is beyond me.
But with enough hype and snazzy naming it cannot help but sell high and wide.


Cheers George

georgehifi,

I never set out to write long winded posts but I admit they tend to be overly verbose. I’m going to strive for more brevity starting now.

Your last post goes over information already discussed on this thread.and totally avoids answering both relevant questions concerning your theory asked of you in my admittedly long winded prior post:

1. Does any scientific evidence exist that in any way supports your theory that the current class D switching frequencies are too low and result in sonic anomalies that are audible?

2. Have you personally ever heard these sonic anomalies listening to good class D? If so, please explain what these theoretical sonic anomalies sound like. Apparently, you currently may be the lone known human believing in, and claiming the audibility of, these sonic gremlins.

Tim

I've always admitted Class-D is the future for hi-end, just not yet.
Very long winded post, but still can’t see the forest through the trees.
Read again and try to let it sink in.
https://forum.audiogon.com/posts/1417196

Cheers George

georgehifi,

Over six hundred years ago, adventurous explorers were looking for financial backing to sail west on the Atlantic from Europe to see what was out there and ,hopefully, discover new lands. At the time, early explorers limited their explorations to sailing around the coast of Africa always keeping the land in sight and never venturing out to open uncharted waters.
This is why these first explorers didn’t discover much beyond the trade routes from Europe to the East and back. The reason they were reluctant to venture out to uncharted waters and truly discover new things was because of the ’Flat Earth Theory’. The Catholic church was very influential around this time and it, along with other misguided opponents of scientific discovery, propagated the ancient unproven theory that the earth was flat and, if you sailed too far west, your ship and everything in it would eventually reach earth’s edge and fall off into God only knows what.
I mentioned the above because your ’Low Class D Switching Frequency Theory’ reminds me of the ’Flat Earth Theory’ in many ways:

1.Just as the ’Flat Earth Theory’ (FET) postponed some early explorers from discovering new lands, your ’Low Class D Switching Frequency Theory’ (LCDSFT) could possibly postpone some of the more gullible audio explorers among us from discovering new audio fidelity in their systems.

2.Just as the FET promotion kept explorers close to the safety of the shore for awhile, your LCDSFT promotion could keep some of the more gullible audio explorers among us close to the safety of traditional linear amps for awhile.

3.Just as the FET proponents used repeated claims of the false affects of an unproven theory to mysteriously prevent the advancement of scientific discovery, you use repeated claims of the false affects of your unproven LCDSFT to mysteriously prevent the advancement of system discovery.

4.There’s obviously more allegorical meat on this bone but I’m stuffed.

You never responded to my earlier questions, possibly from this or another thread, about whether there’s any scientific basis for your theory that current class D switching frequencies are too low and negatively affect the sound of class D amps in the audible range. You also failed to respond about whether or not you have personally heard any sonic anomalies in the sound of class D amps you’ve listened to.and, if so, whether you could describe what these sonic anomalies sound like.

When I first read about your theory, I perhaps foolishly listened to several very familiar and very high resolution (96kHz/24 bit) music files I have trying, as best I know how, to be hyper-critical of the sound quality in an effort to hear if there was any subjective evidence I could actually detect to determine whether your theory had any merit.
Well, after a few hours of failing to detect even a hint of a whiff of a scintilla of even a possible twinkle in the eye of even a single sonic anomaly, I decided it was safe to relax in the knowledge that my class D amps had no sonic anomalies in the audible range, at least that I could subjectively detect with my possibly tin ears.
Subsequently, I searched the internet for any scientific or even any subjective or anecdotal evidence in support of your LCDSFT and discovered there is zero scientific evidence to support your theory and the only subjective or anecdotal evidence I discovered was numerous google references of your thread postings here and on various other audio forums.
I find it hard to fathom why you have repeatedly made false claims of the false affects of an unproven theory on Audiogon and other audio forums.about class D amps. I tend to doubt you do it for some twisted personal version of fun and believe there must be some hidden motivation for someone to so diligently,consistently and repeatedly put forth the time and effort required, as you have, unless they personally judge achieving their mysterious goal as sufficiently motivating.

Don’t worry George Hi-fi, I doubt anyone expects you to actually reveal your hidden motivation or agenda.
Fortunately for current and future class D amp owners, repeated suggestions of non-existent sonic anomalies produced by their amps are not audible to any human I’m aware of.
Even in the infinitesimally small chance you’re theory actually has an iota of merit, the truth is if the thousands (millions?) of current class D amp users cannot hear these elusive, currently not to have been heard by human ears. very suspect and apparently extremely quiet sonic anomalies then they, by definition, do not exist.

Class D=Trash, my Ash!

Later,
Tim

I use a Class D Classe Sigma AMP2 which works very well with my Audible Illusions L3A line stage preamp driving Martin Logan Ethos speakers. The Sigma seems to be very neutral to the music and inputs no signature sound. If more power is a requirement one can audition the Classe CA-D200. For me the sound is best when used with balanced ICs.

I've heard Dusty's (Chanel Islands Audio) mono blocks a few times and was impressed by the sound. Very dynamic and musical. Most impressive was the power to weight and cost ration. Very affordable. 

http://www.ciaudio.com/

Not my cup of tea, as I much prefer the added realism provided by good tube gear. 

A good source tells me that Dusty has a giant killer phono stage about to be unleashed for under $1000.  I'd like to compare it with my ARC PH8 and write a review afterwards.

Frank

May not be trash for long madavid0.


Class-D that can finally equal or better hi-end linear amplifiers, tube or S/S.

 It's closer than you think.

Why We'll Soon Be Living In A Class D World 

"In fact, that transistor technology is available today and is increasingly being used by manufacturers to create near perfect sound quality for Class D audio systems. The greater switching speed of Efficient Power Conversion's (EPC's) eGaN® FETs allow amplifier designers to increase PWM switching frequencies, reduce dead-time, and drastically reduce feedback; in turn, producing a sound quality previously limited to large, complex, heavy Class A amplifier systems. Further, this innovative high-speed switching technology has already disrupted myriad other industries, including telecommunications, medical, and automotive to name a few.  
A high-definition eGaN FET-based system with higher PWM switching frequency, reduced feedback, and higher bandwidth produces the sound that has the warmth and sonic quality that audiophiles demand; while actually improving upon the power efficiency of traditional Class D. Class A audio's historic lesser child Class D is coming of age with eGaN technology. And in the next decade, these systems will replace Class A technology, as well as the silicon MOSFET Class D systems in active use today. 

Significantly, audio manufacturers are taking notice and incorporating GaN-based Class D FETs in their systems. This is why, in another two or three years you will start seeing a slew of new Class D amps - ones based on GaN - rolling out to the market; and, for such varied uses as home theatre, car, boat, portable wireless speakers, along with high-fidelity home systems."  

Cheers George

Anybody here running or have auditioned Auralic Merak monoblocks?  Just purchased a couple and should be arriving today.  Moving to these from a Parasound Halo A23

seanheis1
  I started a thread about it and it appears that the switching frequencies aren't currently high enough to avoid degrading the signal. If it was a simple break in issue, they would be broken in at the factory.

Don't give up on Class D though. When the switching frequencies improve, they will become the solid state amps of choice. Some folks don't hear what we hear and they are fortunate to enjoy these small and efficient power houses. More power to them.

Well said the day will come with much higher switching frequencies, as I also posted to Randy's question

randy-11I always thought the gentle filter slope was b/c steep "brick wall" filtering caused caused audible distortions...

It’s because a simple low order output filter can take the amps full power, but it’s effects reach down into the audio band and still leave some switching noise left overs, hence the need to take it up much higher as Technics did with far higher switching frequencies, so they can be effectively removed, without effecting the audio band.

These days when Stereophile tests a Class-d amp, they put on an external output filter, the Audio Precision’s AP0025 filter, which has a -50db rolloff after the audio band so the 1khz square waves look half decent without the switching noise embedded right across it, (good for sales) 10khz square wave still looks a mangled mess though, trouble is this AP filter can only take very low power, would be real nice to leave it in to listen to, but it would blow up in a micro second.

http://www.stereophile.com/content/class%C3%A9-sigma-2200i-integrated-amplifier-measurements#31YFAPfYVDGeowzZ.97  
And yes steeper filters as you said do have their own set of problems re sound, as ML found out with their No.53 monoblocks.

Cheers George 

Hi atmasphere,

     Proprietary power modules and supply?  Wow!  Very cool and very interesting.


Good luck,
   Tim

  I'm very interested in checking out your take on the first Atmasphere class D amp.
     Too early to share any insider details like modules chosen. type of power supply adopted, any customized components, target price and expected release date?  

We're doing our own circuit, so no modules, at this point we are using traditional power supplies.

Hi atmasphere,

      I'm very interested in checking out your take on the first Atmasphere class D amp.
     Too early to share any insider details like modules chosen. type of power supply adopted, any customized components, target price and expected release date?  

     As to the Audiophile Review article describing class D amps as 'digital' amps, I think the author, Skip Taylor,was a bit inconsistent in the entire article. I think you're right, he was  all over the place on terminology and his knowledge base seemed outdated:

.1. He generically states "Class D Audio amplifiers" which is very close to the name of a class D company, Class D Audio.

 2. He talks about "Class D amplifiers" having to use high levels of negative feedback to compensate for poor open loop performance and meet target levels of TIM, not seeming to realize that the inventor of many recent class D breakthroughs such as UcD and Hypex NCore power modules, Bruno Putzeys, believes there's no such thing as too much feedback in his amp designs and considers TIM to be an irrelevant and meaningless measurement as far as class D amps are concerned. 

 3. The author states:"   By definition, large amounts of feedback introduce transient intermodulation distortion (TIM), which introduces a 'harshness' that hides the rich subtleties and color of the music that were intended for the listening experience."  
       His concept of feedback and TIM have more relevance for traditional linear amps than class D amps as far as 'harshness' is concerned and he doesn't appear to understand that high levels of feedback at all levels of the audible frequency spectrum are the key concept /tool that allow Hypex Ncore modules to sound so good over the entire spectrum, including a lack of 'harshness' and its high levels of detail that enables the rich subtleties and color of music to be reproduced so well.

4. I think the author's 'digital' experience is not primarily in home audio and it was reflected in his article.  
     His discussion about keeping the signal in digital format as long as possible, without converting it back and forth between analog and digital, is a concept I remember from the earlier days of class D when the confusion between switching and 'digital' amps seemed to begin.

     I think it's a good concept that kind of lost focus once it became clear that the adoption of the other system components that could make this concept a reality lagged considerably behind the adoption of class D switching amplifiers.  Components such as completely digital sources and preamps, that would allow digital signals to be inputted into the audio chain, remain in the digital domain while being routed by a purely digital preamp and sent along to a class D switching amp still in the digital domain, where it would be converted from a digital signal to an analog signal before being amplified and sent along to the speakers.

     Most of the early examples of class D amps were actually switching amps that were interchangeable with traditional linear amps but could easily accommodate the above with just the addition of digital inputs.

    The author made absolutely no mention of any of this.
     However, I should point out that, while I think keeping an audio signal in the digital domain as long as possible would likely be a good thing sound wise and with the possibility of enhanced functionality, I believe it also creates a boatload of issues that could be fairly complex, challenge manufacturers' flexibility and budgets, would require industry wide agreement/guidelines and negatively affect the number of customers willing to adopt this new technology/paradigm.   At this point, it may be more trouble than it's worth.
Tim                  , 

Ahhh, makes sense, and yes indeed it does. :)

Thanks for the clarification.

No but I believe from what others have said, the output filter in some Class-D’s is included into it’s global loop aiding it’s effectiveness? that's why I bought it up, just to cover all bases.

Cheers George

Hi @georgehifi

I'm not sure that feedback in a digital amp works the same way, or has the same consequences as it does in a linear amplifier.

Do you have any data that indicates feedback is anything but a positive in a digital amplifier?


Best,


E

A high-definition eGaN FET-based system with higher PWM switching frequency, reduced feedback, and higher bandwidth produces the sound that has the warmth and sonic quality that audiophiles demand;

For a quality Class D audio solution, it is possible to avoid these transitions all the way through the system to the final Output Filter

The key to great Class-d, is to get it’s LOW ORDER output filter to do it’s job of getting rid of ALL switching noise from the audio band before it get sent to the speakers, and without creating any phase shift.
The only way to do this successfully is to raise the switching frequency noise much higher so this low order output filter (without masses of feedback) can do it’s job properly without any residue effects or switching noise remnants being left within the audio band.

This is why Technics have led the way with the new SE-R1, have striven to DOUBLED the switching frequency with these new eGan Fet transistors, to get closer to the ideal above results.

Credit to Mark Levinson a couple of years ago tried to solve this problem with todays switching frequency, by using much lager (to take the amps power) higher order output filters and less feedback, but it was a bit of a flop, with their very expensive. (see the size of the 4 x chokes for the higher order output filter for each monoblock).

http://www.stereophile.com/content/mark-levinson-no53-reference-monoblock-power-
amplifier#rJEmkELOsfXsME4b.97
It’s 10khz!! (not 1khz) square wave without!!! the AP filter, is the best ever seen for a class-d amp (almost as good as a linear amp) because of it’s massive higher order output filter and less feeback, but this high order filter obviously created other problems, and was not received well, that’s why low order with low feedback is the way to go, but it must do it’s job properly, and the only way to do that is to raise the switching frequency noise much much higher.

Cheers George

" Interestingly Primare i32 intergrated amp,that is a Class D design, and got very good reviews here in Europe since its release 5 or 6 years ago, will probably be replaced with a new model later this month."

No wonder why my humble bid was accepted. I wish the damn thing would arrive, already, so I could quit reading and get back to listening.

Maybe we have to get use to the class D amp, as it has a very black background, which makes the instrument and voices sound very clear.
Interestingly Primare i32 intergrated amp,that is a Class D design, and got very good reviews here in Europe since its release 5 or 6 years ago, will probably be replaced with a new model later this month.
Primare have a long product cycle, so it will be interesting to see if they have advanced the Class D design, or gone back to a/b solid state.

 Here's a link to what I was referring to in my last post about reading of faster switching mosfets:

http://audiophilereview.com/cd-dac-digital/why-well-soon-be-living-in-a-class-d-world.html

Hmm. I'm a fan of class D amps (really expect them to bring home the bacon someday and am working on one myself) but not the 'digital' amps to which this article refers, because they don't exist. I expect the author really meant 'class D' or 'switching' when he must have accidentally used 'digital amps'... ya think?

georgehifi,

     From the comments of your fellow mates on www.stereonet.au,
it seems you've been posting about the need for an increase in the switching frequency of class D amps to the 3-5mHz range for years to the point of their apparent consternation.  

     I'm not certain, but I'm fairly sure you've been stalking threads concerning class D, and posting similar messages about the dire need for class D amps to raise its switching frequency to 3-5mHz, for years here on Audiogon.  If it hasn't been years, then I apologize, but it sure seems like years to me and I'd hazard to guess to other frequent readers of class D related Audiogon threads.

     Please excuse me while I flash my 'Literary License Badge' and briefly pontificate: Georgehifi is like a stealthy panther, desperately hungry for the definitive answer to the eternal delicious question of the optimal switching frequency for class D amplification that, due to his being a magnificent beast of nature possessing extraordinary feline aural senses,  he alone having the capacity to ever so faintly detect when that frequency is set a few kHz too low.  
     He persistently lurks and stalks through the wild plains of the audio forum hunting grounds until he spots a class D thread and then he suddenly pounces with an uncontrolled fury upon his allegorical prey and maliciously metes out his vengeance by predictably, mundanely, and rather annoyingly, launching into another repetitive attack/statement concerning his bloodthirsty quest for the elusive optimal class D switching frequency.  

     Okay, I think we've all had our fill of whatever that was.  

      I'm apparently to the point of consternation myself.  Unfortunately, I  must admit I lack the complete expertise of  mathematics, electrical engineering and class D amp circuitry and design that I believe may be required to definitively determine whether your claims of the necessity of a higher switching frequency utilization in class D amps is justified.

     Therefore, the only current recourse that came to mind is to directly quote the acknowledged guru of modern class D amplification as well as the inventor of class D UcD and Hypex NCore technology, Bruno Putzeys.  The quote below is a relevant portion of an interview conducted by Peter Roth (PR) of Bruno Putzeys (BP) done on March 1st,2014 and posted on www.sounstage!ultra.com (I'll link the entire interview at the end of my post):

"PR: At Hypex, obviously, you continued development and implementation of the UcD modules for OEM clients -- but tell me about the much newer NCore class-D devices. Is NCore a further extension of what you’ve been doing with UcD, or is it an entirely different class-D scheme?

"BP: I think it would be fair to say that it builds on from UcD. Certainly the fundamental math is the same. The really crucial part of NCore was to figure out how to improve the loop gain even further from what we had -- UcD had substantial loop gain up to 20kHz of 35dB or something -- and I wanted to go beyond that. As I said, there is no such thing as too much feedback, so I was looking for a way to add 20 more dB. That’s actually a very hard problem to crack, because once you start doing that, you have to remember that a class-D amplifier has a limited bandwidth. A reasonable switching frequency for a class-D amplifier is just under 500kHz or so. If you go much above that, you run into efficiency and headroom problems. That, in turn, implies that you have no more than some 200kHz of bandwidth to play with -- actually, less than that. And if you want to cram 50 or 60dB of audioband loop gain into that bandwidth, you have to think completely differently from the way that linear amplifiers are usually designed.

Linear amplifiers typically have what we call single-pole compensation; some of them have two-pole compensation, but nothing much beyond that. UcD has four-pole compensation, and NCore has five. Once you reach into the four- and five-pole compensation, you have this problem that the amplifier can be operating in perfect stability until you clip it, and then it will suddenly start oscillating at a frequency that will immediately damage the amplifier and the loudspeaker -- so you want to avoid it like the plague. And not just that, but you actually want to return the amplifier, once it comes out of clip, to its normal operating regime so quickly that you don’t hear any glitches. The whole NCore patent revolves around the practical solution to that stability problem, the way that it actually catches the feedback loop at the moment that it’s thinking of going unstable, and then lets it go when it is safe to do so.

Apart from that, of course, I did learn some more tricks as concerns driving the output stage. If you have 50dB of feedback, and you are aiming for -100dB of distortion, you’ll still need to manage to get -50dB of open-loop distortion. In that respect, you are right in your previous question: You have to start out with something that’s good, because there is always a trade-off. If you’ve got yourself 50 or 60dB of negative feedback, but if you can also get 10dB improvement in open-loop distortion, why not do so? So the actual power circuit has also changed between UcD and NCore, and obviously, then, the later UcD modules actually use an improved power stage that was borrowed from NCore. The core of NCore is the feedback circuit, but the actual product contains some more improvements that are now trickling back into the UcD range."

     georgehifi, Putzeys states that a switching frequency of 500kHz or so is reasonable and you're claiming 3-5mHz is necessary for optimal class D performance.
      I'm not actually claiming that he's right and you're wrong.  For all I know, he may believe 500kHz or so is reasonable but may agree with you that 3-5mHz is optimal.  I just don't know if my class D monos would sound better with a higher switching frequency since there's no method to currently compare unless I compare my amps to the $30K Technics amp with the new eGaN FETS.  It's honestly hard for me to imagine how my amps could sound better but I'll be keeping an open mind until I can try a more reasonably priced alternative to the Technics.
      I'm thoroughly pleased with my experiences with class D amps so far and likely a permanent member of the class D Fan Club;I've already got my official club card, big club ring with the diamond studded 'D', club jacket, beanie, "Class D Stands for Delicious" t-shirt, "Class D or Die" bumper stickers , matching underpants and, due to years of practice, now have the unbelievably elaborate secret member handshake down pat.  

     Anyway, my main point is that I know Bruno is a friendly and accessible guy via email and will answer your email inquiries on class D switching frequencies.  Wouldn't you rather receive answers straight from the horse's mouth than constantly raise the subject on Audiogon Forums?  Speaking for myself, I know I'd sincerely enjoy not hearing about this subject again until you, hopefully, post on what you discovered in your discussion with Bruno.  


Complete Putzey Soundstage interview:
http://www.soundstageultra.com/index.php/features-menu/general-interest-interviews-menu/455-searchin...


Thank you,
   Tim  

Tim,

"Hi Ho".  After I realized how goofy that sounded, I told him his  new American name would be "Henry" .  LOL! That is very funny. 

Speaking of which there are a pair of the Nord monoblocks NC500 for sale right now on Agon. Fund is very limited right now, otherwise I would pull the trigger on this just to find out how it sounds compare to the NC400 that I have. Thanks again for reading all of my rambling.

Henry

So, you’re saying it may be a year or two before you buy your first class D amp?
If you’re going to wait until class D is absolutely perfect, the good news is that it’ll probably be the last amp you’ll ever need to buy.
Good strategy.

Yes, I’m waiting till it gets to what I believe will be time for me to turn to the dark side, with a 3-5mHz switching frequency, then it will address the problems I think it has "trying to filter out ALL of it" without effecting the audio bandwidth, and then my linear amps WILL become boat anchors.

One of the tell tale test will be to see in mags an almost square 10khz square wave without any buzz saw ringing across the top "without the use of test AP0025 (Audio Precision) inline filters" that are now being used by mags to hide it when measurements are being done, as I queried here on Stereophile.
http://www.stereophile.com/comment/565267#comment-565267
This then will show good audio bandwidth with total filtering out of the switching frequency.

This is what an unfiltered (by test equipment) square wave looks like on Class-D. If this was a linear s/s or tube amp a tech would say not to use it it has a major problem.
https://www.google.com.au/search?q=Class-D+square+wave&espv=2&source=lnms&tbm=isch&sa=X&ved=0ahUKEwj_g9zOsaDTAhWMnZQKHax5C8kQ_AUIBigB&biw=1255&bih=782#imgrc=OFuO9FVLq5xJeM:

Cheers George

Hi George,

    So, you're saying it may be a year or two before you buy your first class D amp?
     If you're going to wait until class D is absolutely perfect, the good news is that it'll probably be the last amp you'll ever need to buy.  
Good strategy.

Later,
 Tim

noble100658 posts04-13-2017 2:15amgeorgehifi,

Here’s a link to what I was referring to in my last post about reading of faster switching mosfets:

Yes I posted about these before, same guy invented the Mosfet.

https://forum.audiogon.com/posts/1414493

And yes it’s on the way, but this is just a small company able to supply Technics, not the world "yet"
Companies like Motorola will copy it and then the ball will be rolling, but with this transistor the Switching Noise Frequency has been doubled to 1.5mHz but I would like to see it to get to 3-5mHz.
This way the low order filter that has to be used on the output of all class-d amps can do it’s job well away from the audio band, without it’s effects (phase shift) or left over switching noise getting down into the audio band.

Cheers George

Hi Henry,

       You know, if I said "Hi" and used your last name after, I'd be saying:"Hi-Ho".  Mind if I say it twice and add: "it's off to work we go"?
      True story from my college days:
       I worked at a Chinese restaurant and a new guy was hired.  He was a Chinese exchange student whose 1st name was "Ho".  When he came to work the next day, I instinctively said: "Hi Ho".  After I realized how goofy that sounded, I told him his  new American name would be "Henry".  Honest.

       Okay, sorry for my side track.

      You're correct, the Abletec modules have a built-in smps, auto European/North American power sensing circuit and a regulated input power circuit.  D-Sonic also uses a custom input buffer to raise the low input impedance to 60K ohms so it's compatible with almost any preamp; so my former tube VTL preamp (output impedance of 200 ohms) matched very well with the D-Sonic M3-600 monos.

     I believe the Pascal modules also incorporate the smps and other circuits on the same board, so these are likely ruled out for your purposes as well.  At least the Hypex NCore 400 and 500 modules remain as good candidates.  

     So, please knock our socks off with that thing that you do with your analog power supplies!

Thanks for your participation,
              Tim 

Hi Tim,

Thank you very much for your understanding the tip on the Abletec power amp modules.  Unfortunately, I think all modules being offered by Abletec has built-in SMPS and there is not really a way to bypass/separate the SMPS from the amp section.  Even if that is possible, it would not be practical and/or economical.  This is the very reason I have gravitated toward icepower and Ncore so far, as these can be used with power supply of my own design.

I am aware of products being offered by Nord from England.  I believe, and I maybe wrong on this, that there is a misconception about the input buffer in most instances, in that it is choice rather than a necessity.  In reality, I believe it is a necessity first and foremost since most of these amp modules, may have a very low input impedance, in fact even lower than in the case of icepower 8 kohms in most modules, which is very low compared to the industry's standard.  One example would be in the Abletec's case, where if the module is being implement as is, its intrinsic input impedance is less than 2 kohms which is so tough drive properly for any preamp, except a very few preamps, and one being my Fire Preamp.

From what I have heard with the Ncore400, realizing that it is not the Ncore1200 which may sound completely different, Still, I believe that the Ncore technology is most appropriated to be mated with a tube preamp, and maybe preferably, a non-negative feedback design for best result, at least to my ears and in my system.  Thanks

Henry

@pcrhkr - I had a crown xls 1500, which was the first gen if they're more tech. It was a great amp, especially considering it can be had new for not much more than $300. 

I recently bought a used pair of class d monoblocks by Ghent audio.. they are noticeably better than the crown. Just putting that out there.

georgehifi,

     Here's a link to what I was referring to in my last post about reading of faster switching mosfets:

http://audiophilereview.com/cd-dac-digital/why-well-soon-be-living-in-a-class-d-world.html

      This transistor technology is called Gallium Nitride (GaN) and is poised to uproot the high-end audio world.  There's even mention of your favorite subject, higher PWM switching frequency, and how these new GaN FETS will enable it.

Tim

Modern Op amps are very good implemented correctly. Most music is produced with tens of them in the signal chain at least. There's no way to avoid them entirely.

Almost all DAC's rely on them, if not external, then implemented in the filter chips themselves.

Power supply and supporting part quality matters.

Best,

E

georgehifi,

      Yes, Nord does cut costs and prices to customers by selecting op amps for their input boards that they like the sound of rather than the much more expensive method of designing and building their own.
       Class D power module manufacturers typically utilize the more expensive but high quality discrete MOSFET transistors in their output stages.  I've read that there's recently been a lot of effort in the transistor industry going into developing even faster switching MOSFET transistors expressly for use in class D switching amps.  Are class D amps becoming so popular and mainstream that their manufacturers' needs are now influencing the direction of transistor r&d?  I believe the future of class D certainly looks very promising.

Tim

 input buffer boards that allow a choice of op amps
  Maybe this will be the future of class D, with op amp rolling

Can they get any cheaper, in manufacturing with opamps instead of discrete transistors.

Any hi-end linear poweramp that used opamps as the input stage for their amps wouldn't be recognized for very long and die a slow painful death.

Cheers George  

Hi Henry, 

     Thank you for clarifying your findings: "from what I have implemented and heard is that an unregulated analog power supply seem to always sound better than those with SMPS. And this is strictly in the context of class D Icepower."  I understand you're in the midst of experimenting to determine if the same holds true for Hypex NCore 400 modules.

     I'm not really surprised that Hypex didn't allow you access to their top NCore 1200 module. I've read that Hypex had a predetermined sales strategy to  reserve the NCore 1200 module for the larger OEM market while offering the NCore 400 module for sale to the DIY market and amp assembly market.  I believe Hypex is considering H2O a segment of the amp assembly market even though we, and likely many others, realize H2O  actually more closely resembles an OEM than an amp assembler.   
     
     I assume you're aware of the newest Hypex module, the NCore 500, which supposedly is very similar to the sound quality of the NCore 1200 but not quite as powerful.  There is even a U.K. company somewhat similar to H2O, Nord, that offers stereo and mono-block amps based on the NCore 500 module that use Hypex smps but have discrete class A input buffer boards that allow a choice of op amps (Sparkos SS3602 or Sonic Imagery 994).  Maybe this will be the future of class D, with op amp rolling rather than tube rolling.

     There are also many very good alternative class D power modules such as the Anaview/Abletech ALC-1000-1300 (used in the D-Sonic M3-600-M mono-blocks) that utilize Phase Shifting Modulation (PSM) rather than the Pulse Width Modulation (PWM) that Hypex NCore modules utilize and the Pascal modules (used in the D-Sonic M3-1500-M and M3-800-S, Red Dragon S500, Jeff Rowland Continuum S2 integrated and 525 stereo amps) that feature their proprietary and patented UMAC  technology.  I believe these modules would be available to you and could possibly be enhanced with your custom analog power supplies.

Thanks,
  Tim