Good Class D amps ??

[mocenigo]

On 03/04/2018 at 3:12 PM, Ralf11 said:

 

They have a stereo amp under $1,000 USD, correct?

 

Well, the ones based on the NC252MP and NC502MP modules should (I just had a look at the web page) and if you order the Icepower 1200AS2 based one from outside the EU it should be very close to USD 1000 as well.  If you want NC500 + custom buffers, of course they will be way more than USD 1,000...

[mocenigo]

7 hours ago, jmpsmash said:

Has anyone compared the PS Audio Stellar M700 and the Hypex NC400?

 

I had the NC400 for more than a year and I love the detail, neutrality, dynamics and esp how it manage to control the woofer so well. Unfortunately my NC400 just blew (totally my fault) and I am thinking of moving to the M700 but I don't know if it will retain the same qualities. They will be driving a pair of Mezzo Utopias.

 

The PS Audio M700 uses ICEpower 700AS1 modules, which already have a gain of 27.4 Db and an input buffer. So they add an additional buffer to modify the sound, and I guess their "analog cell" is just impedance adaptation, addition of a sound signature, with unity gain. In general the Hypex NC400 and NC500 modules are considered superior to the older ICEpower modules (there are new ones, the 1200AS1and 1200A2 that are preferred by many, and are based on a completely new technology with respect to the rest of ICEpower's product palette). Also, the ICEpower modules integrate power supply and amplification on the same board.

In my humble opinion it makes more sense to ditch the voltage amplification stage of these integrated modules, in other words going from NC400 to NC500, and have a separate power supply (which can reduce interference). Also, adding a buffer to an already complete amplifier makes little sense it is sound character tuning. So it is better to have an input stage that has all the same characteristics as the "analog cell" (class A, discrete, maybe with a modular structure) but also has its own gain, with the result that the signal has to pass through less components.

I would therefore opt instead for a NC500 based amplifier with a buffer stage and replaceable opamps (such as the excellent discrete ones from Sonic Imagery, Sparks, Burson, or DEXA) so that you can also very easily change the sound signature to better match the rest of the equipment. I sourced mine from Apollon Audio, which are also very well built, with a clean organisation inside. But there are also other manufacturers, depending on your geographic location. 

 

I am not affiliated with Apollon Audio - I only have talked a lot to the owner of the company when I was thinking which amp to buy and how to customise it, and am a very satisfied customer.

 

 Roberto

 

[mocenigo]

Just now, GeneZ said:

 

Do you have a 2 ohm speaker?  Know anyone that does?  Most people should not be so concerned.  These amps sound VERY good.  All this data dumping, but no listening?

I am sitting here listening to my ST-10.  Silly fear mongering arguments is all I see.  2 Ohm speaker?   Google it.  Most are car speakers. 

 

And some PA/live concert applications. You can have monsters that go down to 1 Ohm. Then the amplifiers need to provide a LOT of current. Since the amps are then usually in class D, one may be temped to use these products with those speakers...

[mocenigo]

12 hours ago, esldude said:

As I recall the output filtering interacted with variations in speaker loads enough to be audible on the Nuprime.  So it is for real and it would be smart to look at your speakers response and impedance curves. 

 

This happens often with Class D. This is the area where they still have to catch up with some AB designs - for instance the NAP 250 can drive nearly anything.

[mocenigo]

Just now, GeneZ said:

  

Why should anyone need to use an audiophile amplifier on a PA speaker when the cheaper brute units designed for them is all you should need? 

 

In other words?  Its like saying to those so inclined... don't drive a BMW 385i off road.   Because it will get stuck.   ? OH!   I wouldn't buy a BMW then!

 

I have absolutely no idea why  But audiophiles are a strange bunch. And I have seen some DIY projects for serious audio oriented subwoofers that were using 2 Ohm woofers or two 4 Ohm woofers in parallel - where the components actually come from PA applications. Why people should make their lives so difficult escapes me...

 

 Roberto

[mocenigo]

On 11/25/2018 at 6:46 PM, jmpsmash said:

@mocenigo Yes. I agree. I understand the NC400 is for hobbyist with the input stage while the NC500 is the OEM one designed for that exact usage you mentioned. 

 

Yarp. Now, hypex also commercialises a simple input buffer that was initially meant for evaluation purposes only, in order to evaluate the NC500 - but some manufacturers have used that same buffer in finished products that were sold for a few thousand bucks. That buffer is horrible. I believe it contributed, in part, to the reputation of harsh sound and bad highs that in some circles Class D amplifiers still have (Class D does have some limitations - esp going to low and/or very capacitive impedances).

 

A DIYer community developed a modified version to allow opamp rolling, and engineering at companies like Sonic Imagery and Weiss helped by designing better versions. The no-longer very recent wave of amps from Nord, Apollon, and others, use those buffers, and it is in another league. I have Apollon mono blocks, which I chose after carefully evaluating the materials used and the design approach (including the placement and orientation of the PSU, for instance) and I cannot repeat enough how much I am happy with my choice.

[mocenigo]

1 hour ago, Matias said:

 

Also check out the Apollon NC800 SL which is also opamp+NC500 based.

https://www.apollonaudio.com/apollon-hypex-nc500-class-d-amplifier-nc800-sl/

 

I would suggest to take the monoblocks. Two monoblocks cost about 100EUR more (total) than the stereo version and, apart from having better separation, you can place them close to the speakers with short speaker cables and drive them with longer XLR interconnects. Also, the internal placement of components of Apollon monoblocks is a bit better than in the Nord ones. One cannot go wrong with those monos.

[mocenigo]

7 hours ago, jmpsmash said:

 

I ended up getting the monoblock from Nord. They are quite comparable. They are pretty much just the same ideas with the same components in the input stage (Sparko regulator and Sparko/SI op amps). Interesting that you think the Apollon has better layout, the only difference i see is the orientation of the power supply. they have the opposite orientation and the Nord one ended up with shorter cabling from IEC to PS and from PS to the input stage board.

 

The orientation of the power supply matters: it serves to reduce interferences. In particular also the inductors are farther from the amplification circuits and the buffer board. The longer cabling does not matter much because the AC one is quite far (and presumably twisted) and the DC one is not problematic. With this orientation the radiator of the PS also works as shield. Finally, the buffer board has larger spacing between

components, providing better areation (both the class A opamp and the regulators get hot).

 

Maybe not earth shattering differences, but they persuaded me.

 

 Roberto 

[mocenigo]

1 minute ago, Ralf11 said:

 

Can you rephrase this?  especially the part about orientation of the PS

 

I get the impression you are not a native English speaker, so if you are more comfortable in Spanish (or Italian) I am sure someone here can translate it.  Thx.

 

Oh come on, at work I speak and write in English every day in a senior tech leadership position for a major corporation... Maybe I get compliments only because I am in a senior position, but I am not entirely sure. Yes, sometimes my sentences are a bit convoluted, but still - also I tend to break sentences and omit parts (in every language, BTW, including my mother tongues).

 

Let me try to rephrase.

 

The orientation of the PSU matters because of EMI. Inductors, in particular, have strong fields around them. So having them further away from the amplification board may be better. Also, the resulting position of the heatsink (i.e. between the PSU and amplification circuit) will make it partially act as a shield. Wrt the Nord board, the Apollon board has wider spacing around the opamp and between the regulators. This can help with ventilation and reduce the temperature of the boards, potentially increasing life. 

 

Does all of this cause audible differences or improve reliability? I do not know, but it seemed to me that there was some additional attention to detail.

 

 Roberto

 

[mocenigo]

26 minutes ago, Matias said:

 

And I ended up ordering yesterday the Apollon NC800 SL with Sparkos opamp, but I am going to test the SI from a friend as well.

 

I think none of you two will be disappointed. These amps are just great. Everybody knows where my preference lies and I won't repeat it (I also sent them back for an upgrade and I will report).

In order to reveal any "flaws" of good class D amps you need extremely revealing speakers (I think I do but I do not find flaws) and compare them to traditional A/AB amps that cost one order of magnitude more. Especially with very neutral opamps they will  mostly act like a voltage and current magnifying lens of what is upstream.  My preamp is a good chinese "study" of the FM255 (I have to correct the ground wiring...) and I can definitely recognise the sound signature I heard a few times at some trade shows.

 

 Roberto

 

 

 

[mocenigo]

6 minutes ago, jmpsmash said:

 

No takers at the moment. Class D is one of those cases where it is possible to gather some enthusiasts into one location for comparison. The amps can potentially fit into backpacks. Imagine asking ppl to bring in their class A amps, it will be quite an operation!

 

 

It is easy. I bring my 500w monoblocks in my backpack. My friends with class A 500w monoblocks should do the same. No show or death by broken spine means you lose.

[mocenigo]

On 1/18/2019 at 1:45 PM, John769 said:

 

Looking forward to some more discussion. 

 

"The NCore MP modules have an on board PSU and twin channels providing 85-90% of the sound quality of the NC500 in a compact footprint. They easily outperform older UcD amplifiers. They retain all the bass control, dynamics and most of sound stage and resolution of our NC500 amplifiers out performing most commercial amplifiers found on the high street costing 3-4 times more."

 

Anyone listened to the MP variants?  They seem good value..

 

I haven't, not in a 1-1 comparison with the rest of the system being equal,, but one needs to be able to read between the lines to understand at least what the manufacturer believes. They would never say that mids are less transparent and highs a bit harsher 😞 So you can expect that NCore MP based amps will not have the delightful forwardness in the mids and the smoothness of the highs of a NCore NC based amp with a good buffer. To me, saying "90%" of the perceived sound quality cannot make sense at all since we do not know how to measure it, whereas classic measurements would in fact show the two types of amps to be quite similar.

 

 Roberto

 

[mocenigo]

On 1/19/2019 at 1:38 PM, John769 said:

 

What is the SQ difference between the SI 994 and 2x990 Enh Ticha?  Why do Nord only offer the SI 994?

 

I would expect the difference to be quite small, but if your speakers are sufficiently revealing you will probably notice it.  I expect a bit of less noise.  The 990 is essentially one half of the 994 - or a 992 - in a little shielded box, so there may be an even lower noise floor resulting in a even more relaxed and smooth presentation with very sensitive speakers. Also, note that the pins of the "pro" opamps are larger and better spaced, so at the interface they may conduct better than the little pins of the usual opamps. 

Note that Apollon's amplifiers have an astonishingly low noise floor because of the way the buffer is designed. Yes, they are all derived from the same designs by Richard Doporto of Sonic Imagery, with a few custom tweaks, but with these things the devil is in the details, such as proper separation of the grounds, crossing power and signal orthogonally and other details, which look less than perfect in the Bellissimo and Nord designs, and a close look at the buffers in my Apollons shows a lot of care.  That low noise floor means that you could use them with highly sensitive speakers - mine have 96Db/w/m - and hear nearly no noise at all even with your ear glued to the transducers. You can probably use 110Db speakers. Of course this means that either you have a listening room the size of a sports palace or you are using a few hundred mWs (in my case usually 1-2W), but then you are using the class D amp in a region of absurdly low distortion and near perfect linearity. And you still have that reserve in case your music contains some high transient energy loud bass!

 

In my humble opinion the SI opamps sound better than the Sparkos in the Apollon buffer - I have no idea how they behave in the Nord buffers. This with the REV B Apollon input buffer - I had them recently installed in place of the original REV A that I had, and boy they made a difference (also, with the REV A input buffer I was undecided between Sparkos and Sonic Imagery). So if one manufacturer judges that the larger Sparkos (or the larger SI) are not worth the extra expense, they will not include it to reduce confusion.

 

What I would really want to see/listen are the Weiss opamps. Hopefully some of these manufacturers will make them available.

 

 Roberto

 

Quote

Shame nobody seems to have compared them to the higher model.  It's difficult to make a decision..!

 

[mocenigo]

On 1/21/2019 at 12:37 AM, GeneZ said:

Why not cut to the chase? 

 

Someone ought to manufacture an amplifier and call it...   "The .00001 THD."

 

It would eventually gain the nickname...   "Are you Happy Now?  Amp."

 

The Neurochrome 686 amp gets quite close (0.00025%)

[mocenigo]

18 minutes ago, Ralf11 said:

we need a good SET Class D amp with high bandwidth...

 

Single Ended Triode? 😄

 

It is technically possible, I think. After all, a tube is an output device...

[mocenigo]

34 minutes ago, GeneZ said:

There are amps that designers add in even harmonics to produce a very nice "tube" effect on the sound. I have heard it.  It even sounds better than what I heard heard with tubes.  In essence, its the best of both worlds.

 

Because they do this?  The better sounding amp will not show as low distortion measurements.  Who wins? 

 

Well, FM Acoustics pre-amplifiers add a tad of 2nd harmonic to just the mids. Among the most transparent gear in the world, not the most neutral, very nice sound. 

 

 Roberto

[mocenigo]

11 hours ago, Matias said:

Hypex nCore NC1200 and NC2k are now available for OEMs. 👏

 

I thought the NC1200 was already available to OEMs

[mocenigo]

On 2/6/2019 at 11:04 PM, GeneZ said:

 Ahh!  Another good company to be aware of. Thank you.

 

Well, they are good. A bit pricey 😉

 

On 2/6/2019 at 11:04 PM, GeneZ said:

I just noticed that the Resolution Series 255-MKII uses the same type of potentiometer  as found in my passive preamp. 

https://image.jimcdn.com/app/cms/image/transf/none/path/sb99076bb5e16690c/image/ie536af55798bec26/version/1391575986/meticulous-attention-and-precision-in-every-detail-is-one-reason-for-unique-results.jpg

 

A quite common ALPS. The FM acoustics site boasts about custom made components and had selected ones. Who knows how they select the components...

 

On 2/6/2019 at 11:04 PM, GeneZ said:

I gain the second harmonics with my NuPrime ST-10.  Very tube like in a very satisfying way, especially when its obviously sweet and detailed.

 

Interesting that you  get your second harmonics from your class D amplifier. However, I heard very good things about the NuPrime ST-10. My Apollon amps are quite neutral, but the input buffer does indeed give a "class A" flavour.

 

 

On 2/6/2019 at 11:04 PM, GeneZ said:

I also have a good tube preamp that sounds surprisingly almost identical with my passive.  Both use the same potentiometer. Very neutral tube preamp at that.

 

Which tube preamp?

 

On 2/6/2019 at 11:04 PM, GeneZ said:

My passive preamp (when used in balanced mode) is the audiophile secret of the century.   I use only top notch balanced cables and sit back and laugh!  One balanced cable alone costs more than this unit!    Nob sound passive preamp

 

The NobSound will make your high-end system sound as good as the components you choose to implement. maybe I got an unusual piece?  Its so simple in its implementation.  Hefty metal casing.

 

Those things are very good if you have good impedance matching between components. In my case I could connect directly my Soekris 1541 DAC to the monoblocks, but for some reason there is a it of noise. If I put either a similar potentiometer, or this which is even more neutral, then I get no noise. However now I have there an Ebtech Hum Eliminator (the version Wirth XLR) which is simply based on isolation transformers, and thus I have a minimum of components on the signal. The DAC provides attenuation in the digital domain and I put two IMP in line filters before the mono blocks just to have the volume managed at an optimal point.

 

On 2/6/2019 at 11:04 PM, GeneZ said:

I am now learning how power cords can greatly effect what is heard.

 

True, one needs a quite revealing system though to notice it.

 

On 2/6/2019 at 11:04 PM, GeneZ said:

 

I may even place my tube preamp for sale in the near future and concentrate on finding the right power cords.  But, first.  I am going to test to see how different power cords will effect this nice triode preamp.

 

Do not spend too much on power chords. Try assembling some with Supra lorad, or Eupen LPNM industrial cable.

 

 Roberto

[mocenigo]

1 hour ago, ddetaey said:

If you really believe you need 200 hours, it takes approximately 1 full week.

Just decouple your speakers and let you amp do its thing, playing music continuously.

Dirk

 

Without entering in the matter of whether burn in is real - in that case it requires a load. Buy two good 8++ Ohm resistors (100W or more), terminate them with short cables and let the amp work through them at a strong volume. Check the temperature with a thermometer, to avoid burning.

 

 Roberto

[mocenigo]

1 hour ago, Ralf11 said:

"But wait!!  There's more!!!"

 

your speakers are not a purely resistive load, so will the resistor have the same effect on 'burn-in' as the complex impedance presented by the speakers???

 

👻

 

It is just to put the amp with a proper load, to avoid damage with some designs or inactivity (hence no burn-in) in others. 

 

 Roberto

[mocenigo]

27 minutes ago, Ralf11 said:

my point (as crazy-audiophiliac as it sounds) is that if burn-in exists for a Class D design, maybe - just maybe - a purely resistive load will not accomplish burn-in

 

Uh, why? It just has to run.

[mocenigo]

On 6/17/2019 at 4:06 PM, Matias said:

March Audio just introduced their NC1200 based monos for $1,395 usd each. Great price for the nCore flagship!

https://www.marchaudio.net.au/product-page/p701-flagship-mono-block-power-amplifier

 

 

For that price I doubt they bypassed the integrated opamp based buffer with a better one, not to speak provided separated power supplies for opamp and comparator. 

 

 Roberto 

[mocenigo]

6 hours ago, March Audio said:

 

Just stumbled across this.

 

When you say "better" what do you mean or like to see?  The LM4652 is an exceptionally good buffer.  If you feel that a component with 0.00003% THD + Noise (-130dB), IMD of 0.00005%, voltage noise density of 2.7nV etc is somehow lacking, then you need to explain why.

 

Subjectively better. Or with a "preferred" sound signature. It is not a secret that Hypex-based Class D amps became more popular after solutions with better buffer boards and alternative op-amps came out. Maybe it is just because of the buffer boards (the original buffer boards for evaluation were notoriously quit bad, and also noisy). Music reproduction is not only measurement, and the monoblocks I have (with Sonic Imagery 994 opamps, and regulated power supplies) have "converted" many a friend  that thought Class D always "sucked".

 

 Roberto

 

[mocenigo]

On 6/19/2019 at 12:46 PM, Matias said:

No, it uses the standard opamps and regulators that Hypex provides.

 

More info and pics here

 

https://www.audiosciencereview.com/forum/index.php?threads/march-audio-p701-mono-block-power-amp.6862/post-185443

 

Perfect. And it is a good price indeed.

[mocenigo]

7 hours ago, March Audio said:

7 hours ago, March Audio said:

 

 

With respect thats a massive leap.  Its a conclusion without basis.

 

Maybe its just the ineffectiveness of sighted uncontrolled and biased listening and the seemingly inexorable desire of so many audiophiles to fiddle and "hear" benefits. 

 

 

Well, I *did* say subjective.

 

Also, the results of opamp rolling on other circuits prove (scientifically) nothing as well.

 

At the and I bought an amplifier that has great measurements and, most importantly, an absolutely fantastic sound to my ears, and I paid honestly quite little (about 1.2K EUR per channel including shipping).  I even had folks with amps that cost 10x as much come to my place and leave with envy (FWIW).

 

 Roberto