Class "D" amp? I do not have a clue. Do you?

I have been a serious audiophile for 30 years and class "D" audio is new to me. Most important, what do they sound like?
I think my Spectron sounds fantastic!

For what it's worth, I've been an audiophile for 38 years and the Spectron Musician III Mk2 is my first Class 'D' amp.

I agree with Krell man. The Spectron more than likely is the best sounding switching amp on the planet!
They're awesome. I wasn't a believer until I heard them for myself (Bel Canto S300 & Wyred4Sound ST500). I always thought class D would sound very shrill/digital/thin, but that's not the case. They are very powerful, transparent, and smooth.
Krell _man and others, I want you to know that I appreciate your rapid response. I need something that runs cool in the bedroom for the summer.
I have a Spectron and it it sits inside a cabinet...and stays very cool. Previously had a Bryston...which I was told was a cooler running S/S and the Spectron is way cooler. As far as sound Spectron is the best I have had...but then again I have only had the two.
Orpheus10 - I have Rowland Model 102 based on the same Icepower modules as mentioned Bel Canto S300 (and M300). It is very smooth and transparent (nice liquid midrange) but also very revealing. With new speakers (Hyperion HPS-938) I can finally hear smooth natural sounding sibilants. Rowland makes very nice preamp - Capri in the same case, designed as a set.

Here is review of the set:
I'm a fan of Class D for sure. It can be done very well, but not all are created equal. Definitely add Chapter Audio to the list of incredible sounding Class D. It's one of the Class D brands not as well known in North America. Disclosure: I'm a dealer for both Chapter Audio and Bel Canto, so you can take my post with a grain of salt if you wish.
Class D amps are often referred to as "digital amplifiers," which is a misnomer. They are not digital. They are "switching amplifiers." They can be digitally controlled, but most are not.

In Class A amps, the output stages are "on" all the time. This creates a lot of heat, and the efficiency, well, it sucks. They can lose 70% of their output power in the form of heat, which is why you see huge heat sinks on those big Class A solid state amplifiers. Most of the weight is heat sinks. But they're the best in terms of the amplified signal- they introduce the least distortion, and generate an amplified signal of the input wave that most closely matches it.

Class B amplifiers have the output stages conducting 50% of the time- although the non-conducting stage is not actually turned "off" when it's not in cycle, so to speak. More efficient design, but subject to crossover distortion if the transition from one output element to the other is not perfect. Less heat generation, more distortion.

AB amps have the output stages on over 50% of the time. This lessens the crossover distortion, but you'll still get fairly rotten efficiency and a good bit of heat production. The output signal is not an exact replica of the input signal, especially at higher power, when some of the more sophisticated AB amps go from "pure Class A" mode to AB mode.

A Class D amplifier has one output stage fully "off" for half the cycle. This makes for a very efficient amplifier- only about 10% of generated power is lost as heat. These amps use pulse width modulation to take the input signal and generate a square wave analog signal. This is in contrast to the typical sine wave generated by a Class A amplifier. The square wave contains spectral inaccuracies, which is rectified by running this signal through a low pass filter. The resulting signal is a near perfect sine wave amplification of the original input signal.

The result is much higher efficiency. The switching elements are off half the time, so they generate less heat and suck less power than they would in a linear (on all the time) mode.

MOSFETS used to be used as the switching elements, but many of the newer D amps use integrated chip assemblies to act as the switching elements. They are more efficient switching devices than are MOSFETS.

The result? A powerhouse amplifier in a tiny package. You can get a 500 wpc D amp that weighs about 20 pounds. A similar Class A power amp would weigh in somewhere between 60-100 pounds, have huge heat sinks and have to be placed somewhere that they can freely ventilate. The Class D amps generate almost no heat whatsoever, and are designed to be left on all the time. At idle, they draw practically no current at all, so there's no need to power off after you're done listening. Heat dissipation is not an issue, so you get a relatively tiny package that can be placed anywhere you like. They don't need to "breathe."

They sound pretty darn good, too. I prefer tubes, but the Class D amp I own (Rowland 102) is no slouch. It sounds quite good. Weighs about seven pounds. No heat sinks. Minimal hassles. 100 wpc into 8 ohms.

If you're looking for a tiny amp with a huge bang that generates practically zero heat, the Class D amps are for you. Purists may bitch that they're jacking around with the input signal (and they are), which means that somewhere in the path, you have to introduce signal inaccuracies. However, as the technology improves, those inaccuracies are becoming less and less. Run 'em side by side with a Class A solid state amp or a tube amp, and you might find a difference, but it's not much. They don't sound processed or mechanical to my ears. In fact, to me, they sound quite a bit like a tube amp. Not quite as good, but when the 845 amp is heating up to 200+ degrees this summer, and I'm sweating like a pig with the AC on full blast, I may very well switch over to the Rowland and stay cool without sacrificing as much sonic quality as I once thought. I've been using the Rowland more and more lately, and it's getting more difficult to find a huge difference between it and the MastersounD tube amp I use most of the time. Make no mistake, the difference is there, but it's subtle, and something I could definitely live with in the name of comfort. If you live in the northeast or midwest, where air conditioners are more of a luxury that a necessity (as they are here in the deep south), yeah, I can see myself easily switching over to the Rowland during the dog days of summer.
another nod for spectron... great sounding amps (i came from atma-sphere) and a wonderful company to work with (simon is fantastic).
Class D amps also generate lots of RF. At least they used to. That may have changed.
My experience mirrors that of Mrkoven. I have the ST500 and it's not quite tubes, but it also does not have the brittle crystaline quality of many transistor amps. I find the amp entertaining to listen to for hours without aural fatigue.
Hi Orpheus10. I have a Cary A306, not well known, but Class D. For me it's a keeper. Use to own the Wyred 4 Sound ST 1000 which is also quality.
excessive DC offset can burn up some crossovers and/or built in amps. For a more technical explanation, contact Richard Vandersteen. Switching amps are verbotten on Vandersteen products. There are others out there that can be affected. Contact your speaker manufacturer to discuss their product's vulnerability to this problem.
Last year, I owned a pair of Bel Canto REF 1000 monoblocks. I very much wanted to like them,great looks, nice tidy size, low energy, but I gave up after 2 months. Weak bass, nowhere the rated power output - and essentially the midrange sounded as if it was in a tunnel. I am exagerating a bit of course, but the conclusion is that I could never see myself living with those. Life's too short to give another class D amp a second chance in this century.
I have the Bel Canto ref1000 mkii monoblocks. They are the real deal and work very well with tube pre-amps which tend to have higher output impedance (not all Class Ds may be ideal matches for tube pre-amps), and all the speakers in my system, all of which tend to benefit from the high current delivery and damping factor.

For speakers that benefit from this (not all do), they are just what the doctor ordered in a small energy efficient package. For others that do not, well, I expect Class D to be competitive as well but there may also be a lot of viable competition. My amp hunt is over for the foreseeable future.
I'm a fan. But there are substantial differences from model to model and brand to brand. My Bel Canto Ref 1000s sounded good to me but a bit thin and dry. I switched to Kharma MP-150s and the improvement was dramatic. These amps sound very similar to my friends' tube amps.
Have owned several class D amps including tripath and ICE. I prefer the ICE module based amps. They have a natural sound to my ear.
Suggest you pick up a used amp and listen for yourself. The experience will be worth the investment.
I too have had Bel Canto's Ref 1000's in my system. IMO their power and overall presentation seem very well designed when paired with my Eidolon's. In the June / July issue of TAS Nelson Pass gives, what I consider, an experienced and gracious opinion of class D amplification. Unfortunately, like countless other opinions it generalizes these designs as well as assuming the goal of these designs is to achieve "the qualities of good class A amplifiers."

Class D switching, digital, PCM, ICE, etc. amplifiers come in a variety of implementations. Many share some fundamental design features, still, most seem to have their own distinguishable presentation within class D. None of the class D amplifiers I've listened to at any length sounded like a push pull or SET tube, solid state, class A, AB despite others claims.

As a professional Bass player I often play raw digital pre production files as well as analog recorded tapes. What my switching amplifiers do for me is provide an unbelievably revealing and transparent presentation ( what could easily be termed as sterile ) and the most well defined bass. I've found their most unusual quality is their ability to preform at their full output capacity without becoming congested or fatiguing. It's as if they're passing current directly to the drivers. In short they sound like switching amplifiers and their presentation is not for everyone.

My suggestion is to do your own auditioning using robust copper interconnect, speaker, and power cable and an up to date dedicated and shielded AC circuit. Switching to a battery powered preamplifier took my system to another level altogether.
Readers please note that the BEl Canto ref1000 and ref1000m (or mkii as the ref100m has also been called)are two significantly different designs that reprotedly sound different and may work differently in different system configurations. Be sure you know exactly which design is being referred to.

When vendors change a design but only alter the model names slightly, things can be confusing. USing two different designations for the same design (re1000m and ref1000mkii) only makes thnings worse.
Stringreen, please explain your statement. I don't see anywhere on the Vandersteen website or the user manual's where it's stated not to use class D amps.
When I acquired my ref100ms used, the seller kindly obliged me and opened up the boxes to confirm the specific design. I f I were to pay a premium for that model, which I did, I wanted to be certain that the right guts were in there and not just rely on a model label on the rear casing.

Yes, the dc offset deal is news to me. Some explanation of what that is and why it might be an issue with some speakers and class D would be educational.
DC offset can harm your tweeter, depending on the tweeter design and degree of DC Offset

DC offset present in class D amplifiers and all other amplifiers which do not use output capacitors e.g. Plinius.

Before you purchase your amplifier check the specs. Spectron' specs is 10 mV but as a rule its 3 - 4 mV and its all harmless e.g. drives very well Vandersteen speakers.

When discussing different class D amplifiers, my suggestion is to pay more attention to the power supplies which drive it. There are a number of factors influencing class D amp sound We have posted a paper on our web site which hopefully provides some light:

Good Luck,

Every single person will give you a different opin,One, dont believe everything you hear EVEN from me.but for me i couldnt stand them,way way to much bass,,after i listened for a few months , i cant pin it down in audiophile words ,,,but i could not stand them,was the amp system dependent i dont know,but they didnt work for me,

Which one(s) did you try?

I think you are the first person I have heard allude to too much bass as a result of a Class D.

Mine are certainly not lacking in that department, though they sounded relatively lean in comparison to my prior non class D amp at first. I attribute that to the very high damping factor, which works well with my OHM speakers in particular to deliver tight clean and articulate yet powerful bass.

I could see where it might be considered too much in some cases like as a result of some bass heavy room acoustics maybe or other factors upstream that affect bass levels.
"dont believe everything you hear EVEN from me "

EVEN from HIM !

Still, I love my Spectron amp, love its seductive musicality, love its effortless presentation.

Read professional reviews and amp owners reviews and then and only then ask for the amp. Spectron, for example, provides 30 days money back guarantee. I am sure many other manufcturers too.

All The Best
Regarding Class D application and DC offset...I discussed this with Richard personally, and just found a pertinent entry on the Vandersteen the technical section. You have to wade through lots of entries, but you will find the DC offset entry there. It just might be easier to call Vandersteen and speak to him will find him a knowledgeable, and fine person to talk to. I am running Ayre amps which have a protection circuit to turn off the amp if it senses any DC offset.
To much bass could have been a side effect of the Legacy speakers, but i doubt it.Its more likely my amp has a more realistic creation of the sound ,where if its on the disc there will be plenty of it compared to class d where it kinda of overtook the end to the point of distraction.
reveiws are meaningless especially when it comes to these amps,only mag that had it close was hi fi critic
D amps also have much smaller power supplies, so some may lack in the ability to deliver a lot of current. Hence, despite high "power ratings," some may struggle with difficult to drive loads.
According to R. Vandersteen, DC offset is only a problem in their speakers with a powered woofer (the 5,7, and quattro). Using the class d amps in their other speakers does not cause any concerns. I just spoke with him this afternoon about this.
Afc - Ref1000 can drive 2ohm delivering 40A. Size of power supply has nothing to do with it since 2" toroidal transformer at 100kHz can pass as much power as 10" toroidal transformer at 60Hz. In addition voltage doesn't drop since it is line and load regulated. Traditional "linear supply" needs a lot of caps because of voltage drop and necessity to filter out 120Hz (difficult to filter out and audible) vs. SMPS high switching frequency (non-audible and easy to filter out). A lot of caps in parallel have a lot of inductance - not desirable since it is in series with the speaker. SMPS is in reality class D amp (origin of class D) and got bad rap only because of cheap computer power supply implementations.

"A Class D amplifier has one output stage fully "off" for half the cycle"

High efficiency of class D is not caused by switches being OFF. Speaker is always connected by Mosfet switches (there is no OFF) but direction of current changes. Icepower uses single supply and full Mosfet bridge reversing output while Hypex uses half bridge connecting output to positive and negative supply. Efficiency comes from the fact that Mosfets in ON state have very low resistance (in order of 20mohm).

Replacing in class D analog voltage with analog time (duty cycle) allows to get rid of nasty problem of output transistor nonlinearity that in class AB is corrected by deep negative feedback that causes transient intermodulation (TIM) and exaggeration of odd harmonics (unpleasant bright sound).

I don't understand mentioned output voltage offset of class D amps. It has as much offset voltage as any other class amp without output capacitor (it uses feedback/servo to zero output). Self zeroing is probably responsible in my Rowland for 5Hz (and not DC) bottom frequency range.

Another possible contender, that is not like the

"typical" Class D Design; But is able to sit in, with

the Best of the Class A, AB, Designs ANY day. IMHO.

I have the PS Audio "Gain Cell" amplifiers, they are a
"Hybrid" Class D Design.

The GCA-500 is the 500wpc@8ohm and 1000wpc@4ohms

The GCA-250 is 250 wpc.@8ohms and 500wpc.@4ohms

Bass is "Natural", in other words, it is exactly

as the Recording captured it.

Not too deep, or "over-damped", but just right.

The mids. are as smooth as I have heard, with the highs,

not EVER "searing" I own Horn speakers, that are very

sensitive, so low level listening, is mostly how I listen.

These amplifiers, have the uncanny ability to deliver

that very 1st watt, with Authority!

We All, have our favorites, and what is "great" for one,

might be down right, "nasty" to another.

All I can say is that I enjoy Music at a Level I NEVER

thought possible.

Given a set-up that is really "tight", IMHO if you add

either one of the above amplifiers, whether, it it Music,

or Video, the PS Audio, delivers the goods.

*****Pure Enjoyment
*****Forget the Electronics
*****Ability to reach 120db.
*****Lack of Heat

From the PS Audio site...
The GCA 250 is a two channel, 250 watt per channel power amplifier that produces 500 watt per channel into 4 Ohms and drives down to <1.5 Ohms. It has a damping factor that exceeds 1000 and is the amplifier most people will want to own as its 500 watt per channel rating is more than enough for any loudspeaker system we know of, period. The GCA 250 has an input impedance of 47K Ohms, can be remotely turned on/off with a simple 12 volt trigger, is a fully balanced amplifier, and like the GCA 100, it’s ear in the tweeter to hear any hiss out of this amplifier. Distortion levels below 0.1%. The GCA 250 is the amp of choice for any system whether it’s a stereo or multi-channel application. This beauty can power anything.

The GCA 500 is a brute. This is one of the most powerful amplifiers available at any price from any manufacturer in the world. The GCA 500 is a no holds barred stereo, 500 watt per channel power amplifier that produces an incredible 1000 watt per channel, both channels driven into 4 Ohms and is drives down to <1.5 Ohms! It has a damping factor that exceeds 1000 and is the amplifier you will want to choose if you never want to worry about headroom or dynamic capabilities in your system ever again. The GCA 500 has an input impedance of 47K Ohms, can be remotely turned on/off with a simple 12 volt trigger, is a fully balanced amplifier, and like the GCA 100, it’s ear in the tweeter to hear any hiss out of this amplifier. Distortion levels below 0.1%.

The GCA 500 is unequalled in the world of audio amplification. The king, the biggest, baddest power amp around that can play the sweetest violin passage to the loudest rock and roll, dinosaur stomping audio experience this side of the galaxy. If you’ve never witnessed true audio power with limitless headroom and breath sucking performance, this is the one.

They do benefit from a good power cord.

Balanced Interconnects of the Best quality.

I Love Music!
Pugstub - It might be related to fact that Icepower's full bridge places half of supply voltage (30-40V) on both speaker terminals. Non-powered speaker sees only voltage between terminals (difference) but ground referenced powered speaker might not tolerate such common mode voltage. They also mentioned in Stereophile that some electrostats don't work with Icepower for the same reason. Half bridge schemes like Hypex don't have this problem and can be even bridged while Icepower cannot be (already is).
Spectron: DC offset can harm your tweeter, depending on the tweeter design and degree of DC Offset
Simon, can you clarify why that would be? I would expect any reasonable amount of dc offset to be blocked from reaching the tweeter by the capacitor(s) in the speaker's crossover network. Aside perhaps from transients during turn-on and turn-off, or during times when the amount of dc offset might be changing, and in each of those cases I would expect the resulting transients to be minor and brief.

-- Al
"Ref1000 can drive 2ohm delivering 40A "
That means that its peak voltage is 40x2 = 80 volts. Its not bery good!
With medium efficiency speakers the MEASURED peak voltage is above 100v. This why Spectron amps have peak voltage of 240 volts (to cover poor efficiency speakers too). As I understand switching power supplies also generate high frequency noise. I believe these are two the main reasons why high caliber class D amplifiers like Spectron and Mark levinson (No 53) use traditional transformers.
The single exception is Jeff Rowland 312 which uses switching power supplies with PFC (power factor correction) and for its $15k price provide very good sound.

DC Offset: any semi-decent amplifier ( e.g. Spectron I own and enjoy!) has protection circuitry. Richard Vandersteen may have some bad experience with either old design or new "cheap" class D amp. Today its not a problem (if done right)

"With medium efficiency speakers the MEASURED peak voltage is above 100v."

- I don't understand it. What speaker efficiency has to do with the voltage. Icepower 1000ASP voltage is regulated always 80V and since it is bidirectional it becomes +/-80V.

Switching power supplies generate high frequency noise that is easy to filter out and non audible. For this reason Rowland uses switching supply in Capri preamp and he is very noise conscious guy (he used batteries in preamps).
Linear power supply is also switching - current comes in narrow spikes of very high amplitude producing high frequency noise in addition to 120Hz noise (difficult to filter out).

Now - In order to become 1/4 wave antenna for 100kHz SMPS frequency wire/cable has to be 2460ft. Below 1/10 of the wave (984ft) antenna practically doesn't work. Small amount of the amp's carrier frequency (1%) appears on the speaker wires but at 500kHz it needs 197ft of wire to become 1/10 wave antenna.
Kij has it correct. The RDon is very low for MOSFETs. Resistance Device on.
This is a big selling point for discrete power devices. less resistance is less power lost, or rather delivered to the load.....whichever you prefer.

Now, on to the full /half bridge argument.

Above is the Internation Rectifier article on 'd' amps. Very good read. But, this is only part 1 of 2.
The full/half bridge configurations are well covered as are some of the rf issues.

QUESTION: how do the 2 different kilowatt Bel Canto amps differ? Don't they both use the B+O ASP module? Is there only a difference in input circuitry or the Zobel?

Here is a link to the B+O ASP site.

Looking at the technical specs you will see the power output is specified for 60 seconds.....not FTC continuous. The ASP1000 module has only a 30 second time limit.

I like my 'd' amp. No question about it getting the job done.
Kij, How DID you know my speaker wires were 197 feet long? Just kidding!

As for the PS switching noise at the output, that is removed with the low pass Zobel. However, there are special filters used by the measurement crowd which test equipment. Some problems occur when measuring 'd' amps.
"Switching power supplies generate high frequency noise that is easy to filter out and non audible. For this reason Rowland uses switching supply in Capri preamp and he is very noise conscious guy (he used batteries in preamps). "

Hello Kijanki. By chance I owned Capri before I got my "heavenly" full tube Joule-Electra LA-300ME preamp. As strange as it may sound, this tube amp is more silent then Capri.

Next, I spoke with Simon from Spectron yesterday regarding asserion that high-frequency noise (or signal) is inaudible.
He says that then you emit hi-freq test signal it is indeed inaudible. However, then this hi-freq signal is a part of the mixture with other signals i.e. real music then due to the non-linearity it effects human perception of the music and effect listener fatigue.

He agreed with you, however, that future audio equipment will have only (perfected) switching power supplies even in class A ss amps - which will not have a weight of huge transformers and somewhat less of the heatsinks; the same with high power tube amplifiers.

fascinating topic !

"QUESTION: how do the 2 different kilowatt Bel Canto amps differ? Don't they both use the B+O ASP module? Is there only a difference in input circuitry or the Zobel?"

ref1000 is stock Icepower module in a nice box.

ref1000m has additional input buffer and power supply circuits. If you open up the box, the 3 separate circuit boards/modules are clearly visible. It is dead quiet and most nonfatiguing.

I have read that ref1000m effectively addresses RF noise concerns in comparison to ref1000.

I have never owned or heard any other Bel Canto amp so I cannot offer an assessment of sound differences. From what I read however, others sem to concur that there are differences.
Mike - Problem with switching power supplies is that most of them are designed to save money and space and not to improve performance. REF1000 for instance in improved version REF1000m has additional power supply with a lot of capacitors on separate board. I believe that with the same resources (cost and space) better PS can be achieved in switching than linear technologies.

I remember that there was an argument that high frequency can modulate audible frequencies on nonlinear (at this range) transducer - tweeter. Argument is valid under condition that tweeter membrane can move at 500kHz - not likely.

Larger Icepowers have a little lower carrier frequency than small ones and therefore lower -3dB frequency (50kHz vs. 60kHz). It is because stronger Mosfets are slower. Wide frequency range is necessary to provide smallest phase shift in audible range. Even with 60kHz amplifier shows about 30deg phase shift over 5Hz-20kHz.

This is first avalanche of class D amps (remember first SS amps?). Next generations will have faster Mosfets, higher frequency range, better power supplies and better noise filtering. I see it as technology of the future.

Magfan - this Zobel, can be improved as well. In Icepower it is simply common mode choke two filtering capacitors and one resistor. This Zobel network is responsible for time/power limitations and available power at higher frequencies. In practice musical power is only few percent of peak power (unless one listens to sine-waves)and high frequency energy is minimal. Should I pay for larger Zobel toroid to allow high frequency power that would make me deaf? Thanks for the link to IRF primer. I read opinion that full bridge sounds like very good tube amp while half bridge sounds more like high quality class AB amp (less tubey).
I don't know much about class D but I have owned the Bel Canto Ref1000 and they could not drive the bass section of my speakers. I am not saying it is an easy job for any amplifier but I currently have the Spectron mono blocks and they do a fine job. There is no comparison between these amps with my speakers. They are not in the same league period.
Coxhaus - I'm surprised since I know audiogoner who drives with Icepower speakers with nominal impedance of 1 ohm.
Isn't Spectron a class D amp as well? There is probably more to it than just impedance alone.
I do not know the technical details of class D but I have owned the Bel Canto Ref1000. The Ref1000 could not drive my Eggleston Savoy speakers. The bass was lacking and the music was flat. My wife called it 1 dimension. My speakers, I agree are very hard to drive, but I now have the Spectron mono blocks and they quite fine.
The problem with my speakers are they are 8 ohm not low ohms. My speakers suck down as much power as you can feed them. Most amps don't make power until you get into the low ohms.
Then I suspect that you Icepower having huge damping factor at low frequencies (about 4000) overdamped your speakers making very lean bass?

ref1000 (10K icepower stock module input impedance) may not be an optimal mach in terms of input impedance for your tube pre-amp. That might be a factor in weak/poor bass.

ref100m has 100K input impedance specifically to perform better with tube pre-amps.

I seem to recall that newer Spectrons may have higher input impedance for use with tube pre-amps as well, more like ref1000m than ref1000.

Also damping factor could be a factor as well. Not sure about Spectron DF but BC 100m is >1000, which is very high and might produce lean sounding bass with speakers that do not benefit. Or it is possible that this particular flavor of highly damped bass may not be to everyone's particular liking. Then again, what is?

The OHM Walsh speakers in my system do benefit from high damping. The bass is more dynamic, cleaner and articulate than with the 120 w/ch amp I used prior with lower damping. The change was pronounced when I switched amps and sounded relatively lean at first, but that was a temporary illusion.

The difference in the bass with my Dynaudio monitors was significantly less noticeable yet also clearly present.

Bottom line is things sound more real and natural to me compared to live now with the BC ref1000ms in my system than ever before.

Also I do not feel that performance is lacking in any significant area now compared to the best reference systems I have heard recently.

I am always actively listening to establish new reference standards that I can target though, so we'll see....
Mapman - you're right about low input impedance. My Rowland has extra board (with THAT1200 intrumentation amp) to increase impedance to 40k.