Purifi Class D

[PYP]

On 1/11/2021 at 9:19 PM, GUTB said:

Those of you who think you like your class D, I recommend being honest with yourself. Do you REALLY enjoy listening to you music? When was the last time you turned on your class D and got sucked in? Has it ever happened? Is your system spending more and more time left alone and you tell yourself it needs this or that upgrade? Once again I'm forced to accept that class D has serious killer problems and I doubt it's a phenomena I'm alone in experiencing. I don't think it's matter of discernment or having experience with high-end audio.

I don't like my class D, I love it.  Yes, I enjoy listening -- about 6 hours a day.   The sound is, to a great extent, like a live performance.   Class D can have the intimacy, immediacy, weight, richness (for lack of a better word) and dimensionality of tubes.  No familiarity with the model you tried, so have no opinion about that particular model, but Bruno has cracked the code on class D.  And the Mola Mola amps aren't fussy, which I like.  You can just plug them into the wall and they sound great, consistently.

 

Not here to plug my amps or argue, just to say that an overarching statement like "class D sucks and so do you if you can't hear it" isn't valid.  Anyone exploring amps should give class D a try.  Don't fall for the myth that topology is destiny.  

[PYP]

10 hours ago, Iving said:

- whereas making digits musically inoffensive demands money, blood, sweat and tears.

Well, so far I have avoided bloodletting, but the rest applies.  I think your statement was true five years ago or so, but I disagree in the here and now.   Bruno is one of the designers who has changed the digital landscape and not just in amplifiers (his Tambaqui DAC is truly great).  The basic Purifi "engine" is extremely low noise/distortion at a great price.   I assume that the various Purifi iterations sound different because of different input sections used and one needs to find one that suites your setup and tastes.   

 

One well-respected reviewer mentioned that he listens to vinyl and digital for different experiences and that makes sense to me.  The "what's best" approach to audio is silly unless it is "what's best to my ears and mood."  

 

I was never a collector of fine vinyl, so there was no reason to keep my turntable after digital became the dominant recording medium.  But I can understand the equation is different for collectors, especially the folks who have long enjoyed their classical collection.  Or folks for whom digital, even what others think of as great digital, does not agree with their ears.  

[PYP]

27 minutes ago, Iving said:

The convenience argument in favour of digits, notwithstanding, is massive. Not just sitting in chair with remote, but ease of discovery. And I don't (won't!) download or stream.

 

The way I read the thread earlier, Rexp was just having fun grinding his vinyl axe.

Streaming does require more work to get the best sound, but something like Qobuz via Roon is simply put (in my not so humble opinion) a music discovery machine.  And with the higher resolution formats now becoming abundant, great sound is more available more frequently.   Of course, it depends upon the music one loves to listen to and to discover.  

 

A vinyl axe doesn't rust.  I get that part.  But grinding it will just lead to dullness, definitely for the axe and perhaps for the grinder.   Just saying....and having a little fun.  

[PYP]

41 minutes ago, Rexp said:

Which well respected reviewer are you referring to? 

Steve Guttenberg:  

 

[PYP]

2 hours ago, Rexp said:

Recently Steve seems to be 'coming out' as a more pro analogue guy, saying that analogue sounds more realistic than digital, so his rave digital reviews need to be judged accordingly. 

I liked this particular review because it seemed balanced to me, but I don't look for affirmation of listening formats or particular gear in reviews I read. I'm just curious about what people hear, what they value and how they express themselves.  

 

Most reviews talk about how the gear sounds (of course), which is the opposite of what I'm interested in, which is whether I can connect to the music easily and whether the gear facilitates exploring music that I don't otherwise listen to.  If digital can get me there, especially streaming music, that is ideal (for me) since there is so much music available to sample.  I couldn't explore different genres, music played on instruments I haven't heard in person (say, theorbo), different interpretations of the same music, etc., without a streaming service.   Not trying to convince anyone else...

[PYP]

Is the Purifi 1ET7040SA ( High 40A Current Output) for "professionals" only?  I don't see it on the Purify website, only the 1ET400A.   

[PYP]

1 hour ago, Jud said:

 

Back when I built my own DAC, I had the opportunity to try out different capacitors (same published specs) in each of the two different channels. Then I used high quality mono recordings (Pet Sounds) to see if there was any difference between right and left. I thought there plainly was.

 

This was not a blind test, so certainly fallible on that score, but it did have the advantage over the usual A/B testing that I could hear both channels/capacitors simultaneously and did not have to rely on memory. (The scientific consensus based on many experiments is that auditory memory lasts about 4 seconds.)

And one would need to something similar with the two Purifi modules, one of which isn't available to DIYers.  So, there is no way to know how the two modules compare (subjectively).  

 

When Hypex started this way of distributing both DIY and OEM versions of the Bruno-designed modules, the companies that used the higher powered version also tended to use original input circuits, better casework (including better isolation of power supplies and modules), output connectors, power supplies (including original designs), etc.  Therefore, there wasn't a way to do an apples-to-apples comparison.  Does it matter?  Better implementation = better sound.   

 

Perhaps it is more important to celebrate the release of a truly formidable module that is available to DIYers at a very reasonable price.  

[PYP]

9 minutes ago, barrows said:

This could be the case.  A huge part of the training of engineers is that they are always trained to deliver a design which will "work" at the lowest possible cost.  While given design may "work", this approach might not be the best when sound quality is a design goal.  For example, the sound quality of a resistor used in a control circuit for a washing machine is of no consequence, and cost is likely the most important factor besides longevity/reliability.

In my work in audio, when hiring engineers for design help, it has often been hard to get those engineers to understand that cost may not be so important, and that it is OK to spec a resistor which might be 2x or even 4x the cost of a lessor part.

and are parts then tested against a standard?  It seems that to get the sound the designer wants electronic (and non-electronic) parts need to be tested and matched for consistency.   

[PYP]

1 hour ago, Kimo said:

 

Must you continue to prove your worthlessness here over and over?  Maybe, on the 5000th post you will add some value?

Time to Purifi the thread.  Can we move on?

[PYP]

NAD's C298 is based upon the Purifi module.  $2,399 online.  According to NAD:  "Our line inputs, both Balanced and Single-ended, can accommodate all kinds of analog source components by offering ideal input impedance characteristics with linear ultra-low-noise buffer amplifiers to prevent any sonic degradation caused by inappropriate loading of the source device. "

 

They also have the higher priced "Masters" amplifier.  Both amps have a bridged option so that the two stereo amplifiers can be used as mono amps.

 

NAD is clearly a fan of Bruno's designs.  Earlier Masters amps used the Hypex module (which I happily used for years) and at least one of their current DAC+Amp designs uses the Philips UcD module, which I believe Bruno designed when he worked there.    

[PYP]

1 hour ago, mocenigo said:

They are also brutal - no defect in a recording will be masked, which I consider a good thing. 

They also reveal whatever feeds them.  Reducing the noise in the components that are upstream is worthwhile, as is reducing the noise from the electrical system.   

[PYP]

24 minutes ago, Allan F said:

I am pretty sure that the NAD M-series is based on the original Purifi 1ET400A module, not the newer high current 1ET7040SA that is the subject of this discussion.

It appears to be some variation of the 1ET400A -- from the NAD website (bolding added):  "NAD’s renowned designers worked closely with Purifi Audio to customise the Eigentakt implementation to ensure every product featuring the technology would meet the high standards NAD is known for. For this reason, NAD manufactures its own Eigentakt modules rather than buy off-the-shelf."

[PYP]

21 minutes ago, leManu said:

I got a stereo Nilai500 amplifier last week, and I'm pretty disappointed by the sound of it.  Compare to my old (but recapped) Linn Klout, I got more power, the sound is cristal clear, bass is a lot more controlled, but overall it's hollow and thin.  Acoustic guitar sound like the have no casing, just string in on a stick.  Vocal doesn't sound natural and embodied.  The imaging is really poor and the whole thing just ain't musical at all.  I'm wondering if it's just a matter of taste in sound and if all Hypex and new class D sounds like this.  I've been told that I was just enjoying euphonic distortion with my Linn, but I just can't believe it's true.  I've heard the Naim 250 line power amp in my system lately, and while being faster and clearer than my amp, there was still some body to the music.  I also read about those class D amp completely discarding harmonics, which would maybe explain the lack of texture, depth and body to what I'm hearing.  Anyone here can explain what's happening?  Could a pre amp solve a part of my issue or should I just move on from this amp?

play it continuously for one month and let us know what you hear.  These kind of amps take a very long time to break in.  

[PYP]

6 minutes ago, John Hughes said:

Yes, I agree with that. Also, you need to be careful with ancillaries like power cords. 

 and power conditioning.  the power conditioning I had previously for an A/B amp was not the right choice for class D, at least in my experience.  

 

The other quality of a quality class D amp is that it will simply pass through what precedes it (to an astonishing degree).   One of the reviews of my own amplifiers had me scratching my head:  "How could anyone hear THAT from these amps?"  Look to your source(s) for some answers.  

[PYP]

30 minutes ago, barrows said:

Sometimes, the output filter of a class D amplifier can have an adverse reaction to some aspect of a speaker's input impedance, but this is usually not the case with the newer class D designs like Purifi or Nilai.  Do your speakers have anything highly unusual about their impedance curve?  As to break in, yes there is some in my experience, a 200 hour break in period should be entirely adequate to hear all that the amplifier is capable of (My suspicion is that most break in for these amps has to do with the output inductor and capacitors).  It is also important that we never lump all class D designs together!  That would be like saying all class A, or all class A/B amps sound the same.  This thread is about Purifi based amps, and Nilai is another thing-although so far all reports I have heard on the Nilai have been extremely positive.

Otherwise I agree with what has been posted already: current top level class D amplifiers like the Nilai and Purif, really are neutral in and of themselves, and it sounds like your Klout had some extra midrange emphasis/distortion (a coloration).  Perhaps a class D design with a juiced up midrange and some extra distortion would please you more.  Rogue Audio makes class D amplifiers with tube based input stages which might better suit your listening preferences and system. 

Bruno has said that the input stage of his new amp has an outsized influence on the sound, therefore the tube input stage is a very good suggestion as a test.   Agreed that Class D is a topology and that there are many implementations and designs.  My own experience is with a design that requires more than 200 hours to sound its best.  I don't mean to imply that one cannot hear what the amplifier sounds like before that, just that the final sound is more refined.  

[PYP]

8 hours ago, mocenigo said:

 

what do you mean?

 

It is obvious that a properly designed audio component does not need a power conditioner or special power cords: if they do, it is just their own power supply that is badly designed, and this happens often, and intentionally, in many extremely expensive amplifiers, with just a huge transformer and some large capacitors, in order to make then “revealing.” (Please do not argue what I just say, I will not reply because those are facts than anybody with a limited understanding of electrical circuits can verify.)

 

However, a power conditioner having an adverse effect on the sound a class D amp is weird, unless the power supply is switching (and in fact the class of the amplifying circuit is irrelevant), the filter itself is highly inductive or contains ferrites (again, this can be confirmed by designers of such power supplies), and therefore there can be some trouble in the power supply.

 

But if it is a normal hifi or hiend power conditioner, which is usually almost empty and copies the internal of a Schaffer filter, I cannot fathom how this can happen.

 

 

 

if you are speaking about the source material and not the power supply, the most recent Class D designs surpass almost everything else, except for some AB designs which are roughly on par.

I admit to being an Audio Agnostic.  Meaning, I try stuff and see how it sounds.  My observations are personal and highly subjective, of course.  I shouldn't generalize that experience, but do from time to time.  Just trying to report on my own experiments so that others who are trying to troubleshoot an issue might try it too.  

 

A few of the designers I respect most have looked for new measurements to understand subjective listening results.  Bruno Putzeys is one.  He has said that he respects the subjective impressions, but doesn't pay attention to the theories that folks come up with to explain them.  He is a "measurement first" kind of designer and his DAC and amps have brought me listening pleasure for a while now.  His suggestion was to plug these straight into the wall.  I did and they sounded better that way than from my large, heavy isolation transformer at the time.  Years later, I tried a different conditioner and was very pleased with the result so purchased the unit.  

[PYP]

25 minutes ago, barrows said:

I have heard the Kaluga's, with a Tambaqui, and that combo is among the best systems I have ever heard-and I have heard a lot of high end systems.  Of course the Kalugas are highly optimized, being Bruno Putzey's take on the best possible implementation of his Ncore topology for the Mola Mola product.  The Kalugas feature a discrete solid state input stage with Bruno's "single ended/differential" approach, and it appears they power the input stage from a separate linear power supply, rather than from a feed of the main SMPS 1200 (cannot be certain about this).

Basically, the class D Ncore and Purify approaches provide an output current stage which is close to absolutely neutral and without any distortion as we are likely to see, as you mention, the implementation of the input stage is what we are mostly listening to, as long as the speaker is not presenting some kind of really crazy wacked out reactive load.

What approach have you taken with the input stage of your DIY Purifi amps?  

[PYP]

51 minutes ago, barrows said:

I will probably try the Neurochrome boards, as the Neurochrome designer/engineer knows what he is doing very well.

His website is very informative and his pricing is great, especially for the level of production quality that is provided.  And it is cool that he offers consulting services to DIYers at a very reasonable price.  I would think a knowledgeable DIYer could learn a lot from him.  His straightforward explanations on his website indicate a clear (non-dogmatic) view of what is important.  Impressive.  

[PYP]

16 minutes ago, leManu said:

Thank you everyone for your input.

 

I played the amp for 48 hours straight without any obvious change.   I'll continue playing it and keep you posted about any difference.

 

Sadly the Nilai500 doesn't offer any replaceable opamp, so that's not an option. 

 

I feel like one of the main difference in sound that I experience is the lack of decay.  Notes just disappear so quickly that the sound is dry and lack texture, thus warmness.

 

I was never a fan of tubes, but again I was wondering if adding a tube pre amp or external tube buffer would give me back some of the warmth by adding decay to the sound.  And better imaging too.  But maybe I should just sell it and try another Hypex with opamp options...

we all have different preferences, that is why there isn't one type of cereal or toothpaste, etc.  In the end, this may not be the the amp for you and you should be able to sell it quickly (currently they are backordered for months).  

 

At the same time, as we know, these are systems we are building, not individual components, per se.  Would you care to list your other equipment?  Folks might have recommendations based upon that information.  

[PYP]

55 minutes ago, Jud said:

So if what I think is happening is indeed happening (your speakers have a lot of high end response, not so much deep low end, and the Nilai may show you that pretty starkly, more so than the Linn - and perhaps more so than the amp the speakers were tuned with by the manufacturer), then yes, you could play around with pre-amps as a hit-or-miss sort of frequency equalizer. But why not simply solve the problem directly with software frequency equalization?  You wouldn't have to take a hit selling the Nilai, and you'd have a frequency response tailored individually for your speakers and room.

 

You'd need a calibrated measurement mic, which you can get very inexpensively ( https://www.minidsp.com/products/acoustic-measurement/umik-1 ), and from there it's up to you whether you want to learn to do software equalization yourself for free ( https://www.roomeqwizard.com ), or pay someone to do it for you ( for example, https://www.dirac.com/live/ or https://accuratesound.ca/digital-room-correction-drc-calibration/?gclid=EAIaIQobChMI48nRgeWJ_wIVJRh9Ch3P0QWaEAAYASAAEgIOn_D_BwE ).

Getting a good read on the room interaction is a very good suggestion (although experienced ears can give a good idea of problem areas).  Personally, from there I would try some inexpensive acoustic treatments (or build your own if a DIYer and depending upon whether the room is shared-use).  Since you have two amps now, it would be an interesting experiment to try the treatments with both before anchoring them semi-permanently.  This approach does add a lot of variables at once, but these are key relationships (i.e., amp/speaker and speaker/room interfaces) to get right before adding more gear IMO.  

 

Perhaps you mentioned it earlier:  what didn't the Linn give you that you sought to improve?

 

 

[PYP]

11 minutes ago, leManu said:

  I'll start moving my speakers around to see if the change of amplifier could have cause change of speakers/room interaction 

Good that you are willing to experiment!  Placement (including where you sit relative to the speakers) is a huge variable and alone might solve your issues.  Have fun!

[PYP]

interesting comparison of three class d amplifiers.  Comments regarding Purifi vs. Nilai at 1:04:17 and then conclusions section at the end.

 

https://alpha-audio.net/2023/01/january-29-2023-1030-cet-live-stream-review-three-times-class-d-hypex-purifi-icepower/

 

 

[PYP]

2 hours ago, Allan F said:

 

For those interested in the Conclusion, it can be found at 1:39:13

These are just opinions, of course, but it was surprising that at least one reviewer commented on the Nilai vs. the Pass, which is their standard.  If the Nilai gets close to the Pass. especially at the Nilai's price point (OK, you do have to put it together), then it seems like Class D has indeed evolved and improved and that the economy of this kind of product becomes clear.  I don't know if the Nilai mono-blocks would have improved performance, but that would have been an interesting comparison.  

[PYP]

7 minutes ago, barrows said:

IMO,  this test appears to be fine for a comparison within the given system context.  As the input buffers used on the Nilai are a discrete implementation, optimized for the Nilai, I would give the Nilai the theoretical edge there.  The difference with using discrete components vs. "using all in one" IC based circuits, is that a talented designer has much more flexibility in tuning the circuit for the task at hand.  In addition, Bruno Putzeys has mentioned in interviews that the Purifi Eval 1 input stage was not designed to be the best possible input stage for the Purifi module, and that he would do something different if were designing an input stage for the Purifi module if he were trying to make the "best" possible complete amplifier.

To get an idea what the "best" possible, Putzeys designed, input stage might look like, we could consider the input stage on the Mola Mola Kalugas (designed by Putzeys): this is an discrete based input stage, using his "single-ended/differential" circuit approach.  This design takes a balanced input, converts to single ended, and then back to differential to drive the module-in this process it also reduces overall input stage noise by 6 dB.  What approach does Hypex use with its discrete input stage on the Nilai... they do not really say: it might just be a discrete, single, opamp, style circuit, or it might be something else.

I find it a bit disappointing that more manufacturers are not getting a bit more creative with the commercial class D amps they design and sell.  I would like to see more Purifi based amps with unique input stage approaches, using discrete components, and even tubes.  I feel like there are a lot of cookie cutter builds out there, but few really well though out approaches.  It is clear that one can do better than just using a single opamp (circuit, IC or discrete) based approach, at least the DIY Neurochrome buffer is a bit better thought out than just a single IC, but where are the really cool, discrete input stage versions?  

On the Nilai's "newness", it is pretty clear if one believes Hypex's marketing materials on the Nilai, that it is indeed a new design-certainly based on what has gone before, but with refinements and improvements over Ncore.  What those improvements are... well we do not know.  But as with Purifi, it is likely that the improvements are in the control loop.  I am supposing that Hypex did not sit on their hands when the Purifi design was released.

I found the high frequencies are slightly more refined (more extended, and smoother) with Purifi vs Ncore-it is a small thing...  I wonder if Nilai has a similar improvement?

what he said.   My own assumption is that both companies have achieved a new level of lower distortion that is significant and can be heard WHEN the input stage is of a similar quality to the amp module.  

 

Since Bruno Putzeys has made the point that the input stage governs the sound, then differences should be in this implementation.   And, since the final sound is about personal preferences, there is not some kind of objective "best," that determination is made by the individual listener. 

 

As @barrows said, the consumer is best served by companies that do the extra work to design a discrete input stage (as one reviewer cautioned in the video, the amp modules are so good that the temptation will be to provide a box and a markup in price rather than further improvements).  

 

Having had two iterations of the original Ncore previous to the Kalugas, I was mystified by the size of the improvement from DIY, to NAD, to Mola Mola using very similar amp boards.  Until I understood the role of the input stage in the overall sound, it made no sense to me that this kind of improvement was possible (not saying it was the ONLY difference).    

 

This video explains a bit more about the Nilai:  https://www.youtube.com/watch?v=B1exMwmlvRg   

 

 

[PYP]

6 minutes ago, fas42 said:

 

In any audio reproduction chain, any 'character' you hear is due to those parts which are the "worst" in the chain - the point of the exercise is, or should be, is to steadily improve those parts - you end up with, nothing but the sound of the recording itself.

 

The hardest belief for audiophiles to break, it seems, is that getting better sound is an additive process - the reality is, that one subtracts the aspects which degrade the SQ - all fully competent rigs will sound, subjectively, effectively identical ...

My assumption is that even identical equipment (including the same speakers) will sound different in different rooms.   I don't believe there is a way to remove the room entirely from the sound of the overall setup.