restorer-john
Grand Contributor
Go download the service manual and look at the schematic. It's not what you think. And standard architecture they are not.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Hypex nCore vs Class A amps
- Thread starter ppataki
- Start date
Go download the service manual and look at the schematic. It's not what you think. And standard architecture they are not.
I'd rather not (on the road right now), so can you give us a 2-line summary?
HansHolland
Active Member
back to "Hypex nCore vs Class A amps" or Class D vs Class A
Class A has no crossover distortion
Class AB has
Class D?:
Class D switches also between output transistors, but at a much higher frequency than Class AB (above 20 kHz). So the distortion that this switching creates is at an even higher frequency, we cannot hear that!
And yes, feedback reduces the distortion for both the Class AB and the Class D amplifier.
We cannot hear the "crossover" distortion of Class D (already before being reduced by feedback) and Class A has no crossover distortion at all. This fits my experience with (uncontrolled) listining to different amplifiers.
Class A amplifiers are heavy, heaters, expensive and rare. So mostly I prefer Class D amplifiers. Although I must say that there are also a lot of very capable designed Class AB amplifiers.
Class A has no crossover distortion
Class AB has
Class D?:
Class D switches also between output transistors, but at a much higher frequency than Class AB (above 20 kHz). So the distortion that this switching creates is at an even higher frequency, we cannot hear that!
And yes, feedback reduces the distortion for both the Class AB and the Class D amplifier.
We cannot hear the "crossover" distortion of Class D (already before being reduced by feedback) and Class A has no crossover distortion at all. This fits my experience with (uncontrolled) listining to different amplifiers.
Class A amplifiers are heavy, heaters, expensive and rare. So mostly I prefer Class D amplifiers. Although I must say that there are also a lot of very capable designed Class AB amplifiers.
Most class D amps are single-ended, not push-pull, so there is no crossover distortion in the first place (because there is no crossover).
HansHolland
Active Member
you are probably confused by not having balanced outputs (like schematic in post above).
What does this mean?
Is midrange bloom a good thing? Sounds like a form of distortion. I want to hear just midrange. Can we define emotional expression in more precise terms? Not necessarily measurements, but at least in terms of color, macro and microdynanics, etc?
Interesting. I always find live sound to be quite dark with respect to a lot of "audiophile" systems - brightness is one cheap method to give the impression of detail. Any decent amplifier used with compatible speakers has all the detail, not all emphasise it.
True that there is some "cloudiness" or "granularity" in the treble on the NC500, esp if used at high power (for instance if you have low sensitivity speakers), a topic which Bruno Putzeys has recently addressed.
Amplifiers with a lot of distortion emit additional signal-correlated energy. In a stereo recording, these additions to the signal are of course partially independent on the two channels, increasing the separation, a phenomenon which sometimes is perceived as larger soundstage. In fact, low distortion amplifiers might have a smaller soundstage than high distortion amplifiers as a result. The NC500 has vanishingly low distortion. If this distortion is H2/H4 which is quite pleasing to the ear/brain system, then you get a euphonic amplifier with a large soundstage, but is this high fidelity?
Fine, but separate please your impressions from facts. From what you wrote I infer that you are mostly used to something that has a specific "sound" (and since you own your money, you have the right to purchase what pleases you, no quibble) but then you tried something that removes that "sound".
I get this. Maybe you should try an Eigentakt based amplifier. It goes a bit above the NC500.
You only posted simple principal schematics, no real circuit. In fact, class D power amps with higher (= normal) output power have push-pull output stage.
Proper terminology would be more half/full bridge. What both of you posted is half bridge.
Anyhow talking about crossover distortion doesn't make sense for class D since the FETs are always driven saturated to modulate the signal.
Anyhow talking about crossover distortion doesn't make sense for class D since the FETs are always driven saturated to modulate the signal.
Warning: Subjective Content here!
I have not read the entire thread, but I have extensive experience with some class D amps (Icepower ASP 500, Hypex ucD 400, Hypex NC-400 and NC-1200). My NC-400 has been my personal amplifier at home for many years now, it has easily outperformed the other class D amplifiers mentioned (excepting its older brother the NC-1200) in that those amps sounded OK, but seemed a little veiled in the high frequencies: things like bells not ringing out so clearly, and their reverb tails being truncated, cymbals lacking in tonality sometimes (sounding a bit like white noise rather than ringing metal). I hear no such problems with the NC-400.
As to comparisons, I have had a few different Class A/B amps here in the $6K to $9K range, from well known and respected manufacturers, to which I preferred the NC-400 for my system, because of its better detail retrieval and lower noise floor: so that is a win for Class D!
But, I recently had a Bricasti M-25 in my system for awhile (this is an ~$18K amplifier or so), and I did prefer the Bricasti, it is a very, very good amplifier. This Bricasti amplifier is something special. But at around $16K more money, one could probably get a better system by using the NC-400 and spending the left over money on better speakers!
Right now I am working on a Purifi amplifier build, in initial testing these sound better than NC-400 here, and I know they will sound even better with a proper finished build.
I have not read the entire thread, but I have extensive experience with some class D amps (Icepower ASP 500, Hypex ucD 400, Hypex NC-400 and NC-1200). My NC-400 has been my personal amplifier at home for many years now, it has easily outperformed the other class D amplifiers mentioned (excepting its older brother the NC-1200) in that those amps sounded OK, but seemed a little veiled in the high frequencies: things like bells not ringing out so clearly, and their reverb tails being truncated, cymbals lacking in tonality sometimes (sounding a bit like white noise rather than ringing metal). I hear no such problems with the NC-400.
As to comparisons, I have had a few different Class A/B amps here in the $6K to $9K range, from well known and respected manufacturers, to which I preferred the NC-400 for my system, because of its better detail retrieval and lower noise floor: so that is a win for Class D!
But, I recently had a Bricasti M-25 in my system for awhile (this is an ~$18K amplifier or so), and I did prefer the Bricasti, it is a very, very good amplifier. This Bricasti amplifier is something special. But at around $16K more money, one could probably get a better system by using the NC-400 and spending the left over money on better speakers!
Right now I am working on a Purifi amplifier build, in initial testing these sound better than NC-400 here, and I know they will sound even better with a proper finished build.
1) I am curious to read how – subjectively – it compares to the Bricasti and
2) You know that the more beautiful the build, the better the sound! I can confirm it based on personal (and subjective!) experience!
1. The Bricasti, while being nearly the equal in simple aspects like noise floor (both amps had inaudible noise at the listening position) was able to eek out a bit more detail in things like instrumental timbre: such that differences between instruments which sound similar, were more clearly separated and distinguished from each other. The Bricasti also had an ever so slightly more natural tone noticeable specifically on voices (but likely present on instruments as well). I work with Sonore, and we exhibited at RMAF last year with this Bricasti amp and Vivid audio Kaya 45 speakers: this slightly more natural tone made vocal reproduction on the Kayas as clear, and real sounding as I have ever heard (being in the industry I hear a lot of high end systems). I did not get to hear the NC-400 through the Kaya 45s though... These differences were small, but they were big enough to notice, and to matter to an audiophile.
2. Having listened to many, many amplifiers, and many systems throughout my 20 years in the industry and being an audiophile since high school (with an audiophile Uncle and Father), I have learned, mostly, how to avoid allowing things like aesthetics to color my view of sound quality. That being said, certainly the build quality of the Bricasti products is impressive: I have seen audiophile products at the same price level which are not nearly as well built, right down to the quality of the screws used and things such as that; clearly this is an expensive component, but knowing something about what it costs to make things, it is fairly priced. whether it is "worth it" to an individual is a personal decision of course.
One thing I would note is that Bricasti amplifiers are designed to be good performers from an objective perspective, they measure pretty well. Stereophile has measurements of the M28 mono blocks on their site. I am no "objectivist" when it comes to the standard set of measurements, and unlike a lot of people who frequent this forum, I accept that the standard set of measurements does not tell everything about how a given audio component may sound. That does not put me in any "camp", and certainly I believe that the standard set of measurements are important to understand aspects of audio performance, but I feel we have not yet found all the measurements we need to fully describe every aspect of audio performance. I understand that the nature of Amir's site here is to push back against the "magical" claims of some audio products, and some audiophiles, and that is all well and good (and appreciated), but I would like to see a little more open mindedness towards the idea that perhaps what we really should be doing (scientifically) is looking for new measurements to describe the aspects of performance which the current standard set of measurements appear not to be able to do.
By all measures, for example, the NC-400 should sound "perfect" right? I mean, all noise and distortion products are at levels which are considered inaudible. So, how does one listen to the Purifi amp and hear that it sounds "better"? If that is true, we must be missing something with our measures.
If I had money to burn, I would buy one of these Bricasti M25 amps, and be very happy with it. At the same time I would also make a substantial contribution to charity, perhaps to feeding the poor, to somewhat offset the conspicuous consumption.
Anyway, I am currently collecting parts and working on getting a chassis to build up my Purifi amp here. Then I am going to re-cap the NC-400 boards (they have been on 24/7 for about 6 years now...) with some long life Rubycon ZLH caps.
2. Having listened to many, many amplifiers, and many systems throughout my 20 years in the industry and being an audiophile since high school (with an audiophile Uncle and Father), I have learned, mostly, how to avoid allowing things like aesthetics to color my view of sound quality. That being said, certainly the build quality of the Bricasti products is impressive: I have seen audiophile products at the same price level which are not nearly as well built, right down to the quality of the screws used and things such as that; clearly this is an expensive component, but knowing something about what it costs to make things, it is fairly priced. whether it is "worth it" to an individual is a personal decision of course.
One thing I would note is that Bricasti amplifiers are designed to be good performers from an objective perspective, they measure pretty well. Stereophile has measurements of the M28 mono blocks on their site. I am no "objectivist" when it comes to the standard set of measurements, and unlike a lot of people who frequent this forum, I accept that the standard set of measurements does not tell everything about how a given audio component may sound. That does not put me in any "camp", and certainly I believe that the standard set of measurements are important to understand aspects of audio performance, but I feel we have not yet found all the measurements we need to fully describe every aspect of audio performance. I understand that the nature of Amir's site here is to push back against the "magical" claims of some audio products, and some audiophiles, and that is all well and good (and appreciated), but I would like to see a little more open mindedness towards the idea that perhaps what we really should be doing (scientifically) is looking for new measurements to describe the aspects of performance which the current standard set of measurements appear not to be able to do.
By all measures, for example, the NC-400 should sound "perfect" right? I mean, all noise and distortion products are at levels which are considered inaudible. So, how does one listen to the Purifi amp and hear that it sounds "better"? If that is true, we must be missing something with our measures.
If I had money to burn, I would buy one of these Bricasti M25 amps, and be very happy with it. At the same time I would also make a substantial contribution to charity, perhaps to feeding the poor, to somewhat offset the conspicuous consumption.
Anyway, I am currently collecting parts and working on getting a chassis to build up my Purifi amp here. Then I am going to re-cap the NC-400 boards (they have been on 24/7 for about 6 years now...) with some long life Rubycon ZLH caps.
I understand the idea, but honestly: If there would be any measurable factor yet to be discovered... How manufacturers are designing their products, exactly?
Or just that we can't trust subjective and uncontrolled listening impressions, as usual... I mean: What if I claim NC400 sounds better than Purifi, or 1200AS2 sounds better than NC400...? etc etc...
My wife is coming out of the kitchen just to comment on this.Because it is a well known fact that class D amps generally have cleaner and more dynamic bass than class AB amps, right? (whatever "cleaner" and "more dynamic" mean)
That fact can also be verified by pretty much any audiophille wife who hears that difference even from the kitchen.
I did not know this! Wow, I just assumed it was Class D because it's so small/compact.
"I understand the idea, but honestly: If there would be any measurable factor yet to be discovered... How manufacturers are designing their products, exactly?"
I am not sure that "yet to be discovered" is what I am talking about here. I would put this more in terms that the standard set of measures is inadequate.
During the development of the Ncore amplifiers, Bruno Putzeys described his design process. He said he does the maths, and simulates, then he builds a prototype and measures it, then he makes changes until the prototype meets the expectations (measures) developed from the sims/maths. Then he listens to the prototype. If he hears any problems, he then goes back and looks for other measurements which correlate to the problem he found in listening tests. Then he fixes the problem, and starts the same process again. That seems like a very reasonable approach to me.
There often seems to be notion put forth on this site, apparently by folks who actually have no experience working in high end audio, that engineers for high end audio companies do not use measurements in their product development cycle. As if they just are randomly throwing stuff at the wall to see what sticks. Although this is a quaint notion, it is not the reality I have seen in my 20 years on the industry. Every high end company I am aware of engineers their products using a combination of measurements and listening tests, and they all own plenty of measurement gear.
As to the example I posted here, one might be interested in comparing the 10 kHz square wave response of an NC-400 amplifier vs. that of the Bricasti M28 mono blocks (available at stereophile.com). This is one measure where well engineered class A/B amplifiers consistently outperform their Class D counterparts. Perhaps this has something to do with some differences heard during listening comparisons (this is only speculation on my part though, I have not specifically been able to correlate these two things). Of course tube amplifiers tend to do the worst on a 10K square wave, likely due to problems caused by the output transformer.
I am not sure that "yet to be discovered" is what I am talking about here. I would put this more in terms that the standard set of measures is inadequate.
During the development of the Ncore amplifiers, Bruno Putzeys described his design process. He said he does the maths, and simulates, then he builds a prototype and measures it, then he makes changes until the prototype meets the expectations (measures) developed from the sims/maths. Then he listens to the prototype. If he hears any problems, he then goes back and looks for other measurements which correlate to the problem he found in listening tests. Then he fixes the problem, and starts the same process again. That seems like a very reasonable approach to me.
There often seems to be notion put forth on this site, apparently by folks who actually have no experience working in high end audio, that engineers for high end audio companies do not use measurements in their product development cycle. As if they just are randomly throwing stuff at the wall to see what sticks. Although this is a quaint notion, it is not the reality I have seen in my 20 years on the industry. Every high end company I am aware of engineers their products using a combination of measurements and listening tests, and they all own plenty of measurement gear.
As to the example I posted here, one might be interested in comparing the 10 kHz square wave response of an NC-400 amplifier vs. that of the Bricasti M28 mono blocks (available at stereophile.com). This is one measure where well engineered class A/B amplifiers consistently outperform their Class D counterparts. Perhaps this has something to do with some differences heard during listening comparisons (this is only speculation on my part though, I have not specifically been able to correlate these two things). Of course tube amplifiers tend to do the worst on a 10K square wave, likely due to problems caused by the output transformer.
The problem is not having the measurement gear and doing measurements on their products. The problem is that some manufacturers allow horribly measuring gear to pass and go on sale. Totaldac for example comes to mind.
Similar threads
