Jump to content
IGNORED

Good Class D amps ??


Recommended Posts

For starters performance matters more than class of operation.  Each class of operation has its own areas of difficulty.  It really is a poor way to pick what equipment to use. 

 

Below are the graphs at moderate power from Stereophile testing of three well known brands of very powerful amps of class D, A and AB operation.  Which is which? The drive level is at 8 ohms, 4 ohms and 2 ohms with lower impedance giving higher distortion in each case.  We are looking at roughly 20 watts into the 8 ohm load and all amps are capable of well over 200 watts.  This graph is THD+N vs frequency. 

 

 

image.thumb.png.185b393241ae53c7492821e16e12ec52.png

image.thumb.png.4169b85c1689366acdbb8b673d47f1db.png

image.thumb.png.56b51b03c67371fbf1e12882254236e7.png

 

 

 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Just now, crenca said:

 

My wild guess is that the 1st is the class D...or maybe the 3rd...

Come on now, guess which is A and AB also. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment

Anyone care to guess which class this amp operates in from some Soundstage.net measurements.  In this case the load is 8 ohms and each line represents power level.  The green line is 300 watts and the red line 1 watt.  This amps is rated for 350 watts.  

 

image.thumb.png.1ee956469d5e41b86c93b8285647bed7.png

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
2 minutes ago, sandyk said:

 

 Attached is the SC ULD -Mk.3 which has since been superseded and improved on by the Mk.4 version which I don't have a copy of.

Note that the graph here shows THD + N at 100W into 8 ohms.

 The performance is substantially better at lower powers.

 

SC ULD 3 p.3a.jpg

So that is very good, but not quite equal to the Benchmark measurements posted above or the other amp tested by Soundstage.net which btw is the large Bryston mono amp.  The only one to pass  the transparency listening test both sighted and blind by the Swedish AES. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
2 minutes ago, sandyk said:

 But the complete kit, including metalwork for this one is only $749 !!

http://www.altronics.com.au/p/k5165-ultra-low-distortion-mk.3-135w-stereo-amplifier-kit/

 

This old 2011 design can also be further improved by Input and Output device selection !

 An experienced DIY person could markedly improve the measured performance of this amplifier.

 

 

 

Yes, looks like a darn good amp and a super value for what performance you get.  But that wasn't really the topic here.  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
10 hours ago, Ralf11 said:

Which are the best Class D amps?

 

Has anyone compared them with other designs, Class A or AB, etc.?

FYI, I use a Wyred4Sound ST500 in my system.  I do consider it very good.  I haven't said it is the best amp in the world.  But it will play my Soundlabs better than quite a number of A or AB amps.  But that is a very unusual and demanding load for any amp.  My amp is based upon the IcePower modules and the Hypex and such are considered to be improvements on those.  The IcePower Bel cantos are very good class D amps too.  I preferred the Wyred4Sound to them slightly as they do sound a bit different.  The newest Bel Cantos I think better than either IcePower. 

 

Now prior to the W4S I used a large Conrad-Johnson SS power amp.  The W4S is much better than it.  Previous to the CJ I used a Classe 25 which powered the speakers better, but sounded a little less good imo.  I've heard the big Brystons in another system and they are good, but I haven't gotten to hear other amps in that system. 

 

You also have to be careful about what is class A operation.  I owned way back a Classe DR-3  VHC which was 45 watts class A.  What it really was is heavily biased operating a bit beyond 45 watts class A and then swinging over into A/B with what amounted to around 220 watts.  But sure with nearly all speakers you might never ever hear music except via class A operation even with the Quad ESL63s I had at the time.   I had one of the Pass Aleph amps.  Only capable of class A operation.  Sounded pretty nice, but your speakers had to be a real gentle load or it ran out of current and sounded terrible.   I think the Krell KSA50 was only capable of class A operation.  The KSA 250 I think was like the DR-3, around 50 watts class A and then up to 250 or so AB.  

 

A final note, using REW I've measured my amp and speakers at near the amp's rated power with a bit under .3% distortion at 10 khz in room.  So that isn't state of the art, but it is pretty good. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
9 minutes ago, sandyk said:

No.

  You are still trying to show just how good Class D amplifiers can  be in comparison with a decent modern Linear design ! :P

 The SC ULD 3 can also be further audibly improved with a simple front end balancing technique, but you refuse to accept  that any further audible improvement can possibly be heard at such already low distortion values !.

 

http://www.diyaudio.com/forums/solid-state/133018-current-mirror-discussion-15.html

Yes good class D can be very, very good, provide power that few linear amps can match without costing multiples more, etc etc.  

 

Yes, I find it highly unlikely with music any improvement on specs below .1% THD and IMD will be audible.  But with speakers attached and maybe not well matched even otherwise good amps can run afoul of those limits.  So those numbers are thru the speaker as heard.  Your only evidence to the contrary is sighted listening impressions which I find of limited value.  More attention should be paid to distortion heard with speakers attached.  At normal listening levels the differences in distortion you think are audible would be like adding or subtracting an amount of sound that is 30 db below the audibility of human hearing, and with noise in the listening room obscuring it even then. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
17 minutes ago, Nordkapp said:

Bel Canto ref150s, Mcintosh MC202 and 252--all sort of smooth, tonally adequate,but ultimately lacking punchy dynamics and enough detail for my tastes. Now, I once owned an incredible amp designed by Demain Martin and Richard Marsh-Marsh Sound Design. It was a 100 watt AB. That amp was amazing on all levels. I've owned some other tube stuff as well but always found it too billowy and slow. My current CIAs with Hypex Ucd modules tick all the boxes for me. But it all comes down to personal preference. For me they work. That's it. 

Just listening impressions of Bel Canto I think they go for that smooth sound.  I find it obscures detail vs some other amps, and also that the low end seems restrained even though they have good power. I do notice they have higher output impedances than some class D amps.  I wouldn't be surprised if that isn't the reason for that sound, and intentional on their part. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
1 hour ago, Ralf11 said:

I've heard a lot of good things about w4s

 

Wyred 4 Sound ST-500MKII is "only" $1,500

 

Blessed by TAS too!

 

Wyred 4 Sound ST-500 mkII/ST-1000 mkII
$1499–$1999
These high-efficiency power amplifiers—delivering 250 and 500Wpc respectively—have got to be among the great bargains in audio. Although they were not accorded a formal review, PS used them to drive the Emerald Physics CS3.2 II speakers, and then for a good month auditioned them on his Quad 2805s. Throughout they performed flawlessly, delivering sound of exemplary clarity and control, perhaps fractionally on the cool side of neutral, but not excessively so and never compromising the reproduction of warmth. Pricing depends upon options (i.e., WBT binding posts, power, etc.).

 

http://www.theabsolutesound.com/articles/2017-editors-choice-power-amplifiers-1000-3000/

 

 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
On 3/28/2018 at 1:59 PM, Ralf11 said:

from the other thread...

 

apparently anything based on NCore is good & maybe the new ICEPower models

 

 

 

I intended to mention this before now.  Class D amps do have output filtering to remove the switching wave they work upon.  This makes them a little touchy about which speaker they are partnered with and probably why the common complaints are about the treble.  So mine works great with my speakers, but may have some output treble anomalies with other speaker loads.  I tried mine with some nice Audio Physics speakers once and it was not pretty.  Sometimes hard to tell which speakers causes issues.  Early Tripath based class D had a fairly large problem with this.  Great on one speaker horrid on another.  Later IcePower modules were somewhat better, but different companies use their own modifications to the output filtering.  Bel Canto leaves more output impedance than most.  Others use what is stock configurations.  And everything in between.  Ncore has been least effected by this though again various makers may use various filtering schemes.  

 

Here is an example using an early IcePower Bel Canto.  Measurements done by Stereophile.

 

Black line is their simulated loudspeaker load.  Blue is 8 ohms, magneta is 4 ohms and red is 2 ohms.  Obviously they optimized their filter for 4 ohms. 

555Belfig1.jpg

 

Here is Bel Canto's recent Hypex Ncore based amp from 2016.  Colors refer to the same impedances. 

 

916BC600fig1.jpg

 

So if I were going to try an amp now in the price range the OP has mentioned I might try a Nord which uses the Hypex Ncore modules. 

 

 Do notice the upper graph has an expanded scale and the worst curve for the simulated loudspeaker has a maximum variation of about .4 db until you hit 20 khz.  Possibly audible, but not horridly off.  Certainly a variation of the lower graph of maybe .1 db is better.  

 

EDIT to add:

 

Here is a more recent IcePower based result.  Still more variation than Ncore, but less than previously. 

 

218PSM700fig01.jpg

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
41 minutes ago, sandyk said:

 I'm not the one making excuses.The facts remain about the huge increase in distortion figures to 20KHZ with the nCore amplifiers that you refuse to accept can affect transparency.

 I tried to be conciliatory and meet you half way in my recent post, but then you just had to come back with this comment, didn't you ?

 

 I am  out of this fanboi thread. I will leave you to GUTB.

 

Huge increase is relative depending upon your perspective.  

 

Something with .1 or even .2% at 20 khz isn't all that large (and you'll not hear it because all harmonics are ultrasonic).  

 

Huge increase from .1% to 2% is getting up there which you might get in some gear like tube amps.  Or they might even run along at 1% at all frequencies at high power with no increase. 

 

So going from .0003 to .1 % is a huge percentage increase, but still small potatoes. 

 

Staying at 1% is no increase in percentages and yet larger than the "huge" increase of class D amps.  

 

While something stays below .01 % THD at high frequencies is better performance I don't think it is audibly different than one which rises to .1% or manages .001%.  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Just now, sandyk said:

Dennis

 Just briefly, before I leave this Fanboi thread.

If I posted figures like those shown on page 1 of the other thread,for distortion vs. Frequency with my 15W Class A amplifier,

or any other Linear amplifier for that matter, you would just about wet yourself laughing , and tear it to shreds in your reply post !

 Because you own a Class D amplifier you make excuses for the same deficiencies, claiming they can't compromise transparency.

I have yet to previously see you let go of measured deficiencies so obvious with any other type of equipment.

 You would rip the design to shreds and claim that it couldn't possibly work as well as many claimed, possibly quoting Archimago and posters in other forums as well. O.o;):P

 

Alex

Find an instance of me doing such a thing.  I don't think I would on something that hit a tenth or two at 10 khz and above.  I might if it occurred at lower frequencies because they might under some circumstances be audible.  And I have said, it would be better performance if that didn't happen.  Yet I don't think it a real big audible thing.  

 

Closest I can recall is how some tube amps of moderate power quote a bandwidth at 1 watt and 1 % distortion with less bandwidth and more distortion at higher powers.  And yet I also have said I enjoy the sound of those despite those numbers. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Just now, sandyk said:

 

You conveniently ignore your attacks on the USB Regen which had pissy little artifacts at levels you consider are well below audibility .

 Anyway, carry on with your preaching to the converted.:P

I don't think I said the Regen had audible issues.  I think spending money on something promising a cleaner USB signal or cleaner analog results from a USB fed DAC which instead makes the signals worse is not a testament to its worth.  Leave out the Regen and get a cleaner analog signal.  That I might have said. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
18 minutes ago, GUTB said:

 

I heard the Kaluga last year at AXPONA, was unimpressed but I don’t like making definitive judgements in that kind of environment. The statement that a traditional class D topology like the nCore stuff compares with likes of Solution and Constellation sounds like fake news.

I think your fake post intended to compare to Soulution amps.  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
29 minutes ago, GUTB said:

Bruno Putzeys is an experienced engineer who frequently self-promotes by saying everyone else is too dumb to understand whatever it is he's working on at the moment as well as he does. Why are his class D modules so much better than everyone else? Because he's smarter than them. Why is his Kalugas better than the Theta? Because everyone else is ignorant about power supplies.

 

Why does Nelson Pass, who is far more important and a universally respected engineer in audio not act like that?

 

 

 

 

I can see why it seems that way to you.  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
1 hour ago, monteverdi said:

So what about LPS. There is a general idea that toroids are the best transformers. They were developed for military reasons for saving space in airplanes and because they were fancy and and expensive (not so much anymore) they were the things to put in better audio equipment. What about R-core transformers, much less primary to secondary capacitance and therefore less high frequency noise leakage.

I like that forum to learn and not to preach!

 

Not trying to be cute with you, but any of those methods can be done exceedingly well.  They all have issues to be solved, advantages inherent in the design.  But all have to have the devil in the details beaten out of them to work well.  SMPS too. 

 

In LPS you can have capacitor filtered, choke filtered, and actively regulated just to pick a few broad categories before getting to the xfmrs in use.  You can use big xfmrs for good reason and put them outside the main box for the gear to get hum away from it.  You can do what some aircraft one did run everything at 400 hz so the xfmrs can be much smaller.  

 

The bottom line is any of those can work and work well.  If I tell you I have an SMPS it might be state of the art or it might be a wall-wart radiating garbage in all directions.  Same for the xfmrs types in LPS.  You really can't say, "R-core xfmrs are the only basis for great quiet power supplies."  Or EI core or C core or whatever. 

 

This is not a bad basic explanation of the issue here:

 

http://www.audioxpress.com/article/Power-Transformers-for-Audio-Equipment  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
48 minutes ago, DancingSea said:

How about we drop this pointless feud and return to discussing options for good class D amps?

 

Seems simple enough ???

We just can't do it. 

 

I'm stuck with class D.  I mean bad enough already right?  Mine has toroidal power transformers I think.  Bad, bad bad.  Leaking noise everywhere.  I mean the output noise when idling is literally a couple dozen microvolts.  It sounds quiet to me.  Quieter than any amp I have ever owned.  But I am having trouble sleeping at night.  I wake up with nightmares about what might be going on and still think it sounds okay when it just can't be, right?  It just cannot be. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment

Maybe LPS would be better if like switching supplies we move the base frequency out of the audible range.  So we can go low.

 

I propose 1 hz AC especially for audio.  We'll need big transformers and even bigger caps.  But the 20th harmonic of 1 hz just enters the audible range.  We can make t-shirts.  

 

1ACforME

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
3 hours ago, Nordkapp said:

Would you mind briefly outlining your education and background in science or engineering, @GUTB  ??

 

 

Not happening.

I think the text books used in his education were TAS, Stereophile, and HiFiNews & Record Review.  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
1 hour ago, sandyk said:

 

Your usual Objectivity out of the window, and running with the Hounds Dennis ?:o

I also note that recently you are also using a great deal more sarcasm lightly camouflaged as Humour.

You can do much better than that !

Would you like more humor camouflaged as sarcasm? 

 

I did remove the comment you quoted after a couple minutes.  

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
  • 7 months later...
32 minutes ago, crenca said:

 

Not motivated to document it for you (cough).  I recall  one forum where the (or perhaps "a") principal of NuPrime fielded these reports, and did not deny them...

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. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...