300w Pure class A Monoblock search!

[LTig]

If you measure it you can find the critical distance which is where reflected sound energy equals direct sound. You'd find the first doubling from close is a 5-6 db drop. Cricitical distance is where a doubling causes a 3 db drop. Eventually the drop with doubling pretty much dissappears as you are in a reverberate field. I've found in most rooms people end up sitting about at the critical distance. In most rooms it is 3-5 meters from the speakers.

Thanks for the explanation. So now I think that @dlaloum's cakculation is correct for 4 m listening distance.

 

[Rick Sykora]

Still not sure if is clear the OP knows what he needs, but if Class D, a couple of Boxem Purifi monoblocks are just about perfect. If not Boxem, there is Apollon, Nord and Audiophonics that all offer comparable options. If willing to consider US suppliers, Buckeye is the forum favorite.

For anything other than Class D, options are very limited. The closest amp that I would recommend is the Outlaw Audio 2220. May not be as much (rated) power, but nothing comes close to the value at $700 for a pair of monoblocs AFAIK. OP should look closely at the ASR review. Have more amps than I need and still find these tempting.

 

[LTig]

Maybe a pair of used NAD 2200 in bridged mode - they have tons of dynamic power.

 

[Martin]

Parasound Halo A 23+ (Black)

Stereo power amplifier

www.crutchfield.com

And if you really want to spend $1500 pick up a used A21. At 250W per channel it's all you'll ever need.

 

[Thorsten Loesch]

Are there any recent properly conducted ABX listening trials for Class AB vs D?

Never ever have been in the first place, not even AB vs AB.

Audio ABX testing (as per ABX Company of Troy Michigan and Dave Clark et al) is not valid test for audible differences, due to a wide range of errors in methodology and statistics.

It actually has more in common with the shell game than with scientific investigation.

Thor

 

[Thorsten Loesch]

Still not sure if is clear the OP knows what he needs,

Maybe he doesn't.

Presumably his speakers use tweeters (and midranges) using a voice coil would around a steel core.

In this case he should make sure that the carrier breakthrough (in other words switching residue) of the Class D at amplifier rated power is > 60dB (better > 80dB) below the operating level of the tweeter with music. That operating level, even with overall full power is way below power, perhaps 5% of full power.

Hint, a 2nd Order low pass will not get us there, much more attention is needed. Class D Amplifiers should be treated as high power AM Transmitters, not like analogue, non-switching audio Amplifiers.

For Class A or Class AB amplifiers there is no carrier and thus no carrier breakthrough.

Thor

 

[Mart68]

Hi

Never posted before. But I am in the search for a set of pure class a Monoblocks with a wattage output of 300w into 8ohms….. I cannot seem to find a useful site to help my search filters!

Does anyone know of any that don’t cost the earth?

Budgetary speaking in the region of £1500 each.

Hopefully you guys can help me find my endgame amps!!

Thanks
Lew

whilst I don't disagree with all the comments saying you don't need class A or 300 watts (or mono blocks for that matter), if you really do want a high power class A just for the sake of it the Krell KSA100 is a dual-mono design offering 138 watts into 8ohm and available on the secondary market inside of your budget

Krell KST-100 power amplifier Measurements

Sidebar 2: Measurements

www.stereophile.com

Used krell ksa100 for Sale | HifiShark.com

Used Krell Ksa100 for sale on 300+ second hand hifi sites & shops. Use Hifi Shark to monitor pricing and global availability

www.hifishark.com

 

[Thorsten Loesch]

The reason i was aiming towards class A is that i have read a lot of information about them being the 'purest sounding' amps that tend not to colour the source with any magic.

Class A does have some advantages, especially if we take into account problems that are, according to audio objectivists mythical because they do not show up in standard Audio Precision tests.

Specifically, Class A Amplifiers, if they truly operate in Class A are free from a number of distortion mechanisms that plague Class AB/B amplifiers and can be quite audible and objectionable.

In the end all amplifiers have distortion and what matters is how audible this distortion is, as well as if there are additional hidden variables that only actualize in a system (like Eddy current induced distortion causing intermodulation in the tweeter voice coil between the signal and the switching frequency residue).

This would be the easy starting point as i want the amp to sound as natural to the recording as possible. I mean, that's the goal right?

I agree it is my personal goal, but I would not assume everyone else has the same goal.

So are we saying that the difference between classes in inaudible?

It is hard to isolate individual specific technical causes for sonic changes, it's a complex topic and often objective measurements only tell part of the story.

To me, an excellent Class D Amplifier can subjectively sound better than a mediocre Class A amplifier.

Once certain levels of excellence are achieved, sound quality becomes similar. So it is less "what the designer does" (Class A, Class AB, Class D) and "how the designer does it".

The reason for the 300 watts, is to drive some RF-7 floor standers I have my eye on.

The RF-7III seem to be 93dB/2.83V/1m with an impedance that drops below 3 Ohm.

If we go by THX recommended levels we want 105dB @ 0dBFS. As music is usually more correlated that uncorrelated between channels we can expect a stereo pair to output ~ 3-4 dB SPL more than a single speaker. For 4m distance we get a worst case 12dB drop in SPL (less in reality).

So a stereo pair at 4m will produce 84-85dB @ 1W. I would get an amplifier that offers at least 10W in class A into 4 Ohm and 200W 4Ohm per channel minimum.

Others commented on Parasound, most of their amplifiers are John Curl designs and rather good. Second hand they tend to be affordable. I'd look for relatively early production that used lateral MOSFET drivers with Bipolar outputs, this combination has some advantages in terms of thermal behavior.

I'd look for the biggest and most high powered designs with the largest number of output pairs. Each pair tends to be biased at around 100mA (for a range of reasons) and each extra pair of output transistors gives more "class A" so having 5 output pairs gives up to 1A peak on Class A, or 2W Class A into 4Ohm.

An additional option that will allow reduced distortion is to biamplify the Speakers. The HF Amplifier only drives the HF driver and crossover. This means at low frequencies where the bulk of the power for music resides the amplifier operates in effect unloaded. This means all class A power is used only in the treble and overall distortion in the HF region is lowered, even if passive crossovers are used.

Equally, the LF sections lowpass will remove some distortion components from the woofer and they will not be reproduced, while with a single amplifier they will be produced by the tweeter.

So rather than a pair of high power Class A Amplifiers, if you can find them at all, a pair of well designed, medium power class AB , Amplifiers biamplifying your speakers may give better results.

Ages ago Luxman (I think) offered wat was essentially the same amplifier in two versions, one 200W 4R Class AB and one 50W 8R Class A to be used biamplifying passive speakers. Good idea BTW.

While here Class D Amplifiers are the fashion, in my experience and personally I prefer well designed Class AB or Class Amplifiers in most cases.

So don't buy Class D "blind", audition with the actual Speakers you want to use and use "loud, complex" music that challenges especially HF clarity when playing loud. Make sure to compare a decent Class AB Amplifier, level matched. Go with what you like better.

Thor

 

[Lindirizzo]

For real class A:

http://www.amaudio.it/assets/p1x-6x-a300x--audioreview336.pdf

 

[antcollinet]

For real class A:

http://www.amaudio.it/assets/p1x-6x-a300x--audioreview336.pdf

A mere 58500 of your earth Euros. Per channel.

Or for around 6 to 7% of the cost, you could buy 2 Benchmark AHB2, run them bridged, get probably better performance, and more power. (at least into 8ohm)

Or just get a purify stereo amp at <2% of the cost, and still plenty of power.

 

[Prince Valiant]

Yamaha MX-10000 comes close at 250W/8R but is not Monobloc:

https://audio-database.com/YAMAHA/amp/mx-10000.html

Practically guaranteed to be outside their budget if one appears for sale. There are a couple listings up on ebay, from Austria, for over twice the mentioned budget.

 

[Matias]

Maybe he doesn't.

Presumably his speakers use tweeters (and midranges) using a voice coil would around a steel core.

In this case he should make sure that the carrier breakthrough (in other words switching residue) of the Class D at amplifier rated power is > 60dB (better > 80dB) below the operating level of the tweeter with music. That operating level, even with overall full power is way below power, perhaps 5% of full power.

Hint, a 2nd Order low pass will not get us there, much more attention is needed. Class D Amplifiers should be treated as high power AM Transmitters, not like analogue, non-switching audio Amplifiers.

For Class A or Class AB amplifiers there is no carrier and thus no carrier breakthrough.

Thor

Typical switching frequency FUD, I have read tons of them through the years. FAQ #6.

Class D FAQ

This topic is dedicated to Frequently Asked Questions about class D amplifiers. If you have been directed here the reason is that it is easier to concentrate the frequent answers here instead of repeating them in several topics. Q1: Class D power ratings cannot be true. How can these small and...

www.audiosciencereview.com

Have you seen real high frequency distortion artifacts in linear amplifiers, and compared to state of the art class D? See below.

Effect of measuring bandwidth on class D amplifier measured THD and THD+N parameters

Effect of measuring bandwidth on class D amplifier measured THD and THD+N parameters It has been quite long lasting discussion here on measuring bandwidth for THD+N plots vs. amplifier power, measured at various frequencies. Especially for class D amplifiers, THD+N results get rapidly worse, at...

www.audiosciencereview.com

 

[Thorsten Loesch]

Typical switching frequency FUD, I have read tons of them through the years. FAQ #6.

Did you ever measure the results in the current in a HF driver voice coil? I have.

Planar & Ribbon drivers are impervious. I routinely run "MHz" level Class D without filter on these drivers with excellent results. In fact, it was so good that I made a filterless Amp prototype for a product with voice coil tweeters only to find major sound quality issues, that were solved with a filter that provided 110dB carrier suppression (and not more feedback).

Most voice coil based tweeter show significant problems, some more, some less.

Class D is now like Class AB used to be. We are at a point, where by combining the available body of knowledge we can make really good ones, fundamentally free from audible fidelity impairments. It took a long time.

In class D every few years previously "perfect" and over HYPEd Amplifiers are suddenly "X'ed" because a new previously overlooked distortion mechanism makes them obsolete and they are relaunched with more negative feedback as the sole solution (instead of applying appropriate engineering).

There is a lot of objective difference between the output from a good linear Class AB amplifier and that of the best Class D Amplifiers. It is worth considering what these do on a system level, with music, where the carrier breakthrough is variable frequency and variable level but usually greater than the actual signal in the HF voice coil.

Have you seen real high frequency distortion artifacts in linear amplifiers, and compared to state of the art class D?

Yes. Linear Amplifiers HF artifacts depend on the specific design. Just THD & D on the Amplifier output are meaningless, the acoustic output from the drivers are important.

These are directly proportionally strictly and only to the current in the voice coil and NOT reliably directly related to the voltage across the Voice coil (due to magnetic effects, eddy current effect etc.). That has various (but different) consequences for both linear and switching amplifiers.

One major issue is that most people designing speaker drivers barely understand how amplifiers functions on anything but the grossest Makro level and most people designing amplifiers understand how speakers functions on anything but the grossest Makro level.

As a result most amplifiers are designed contrary to the requirements of the speakers and visa versa. This is made even worse by an excessive reliance on objective tests that have no demonstrated reliable correlation with "good sound" in both camp's.

Anyway, if you don't hear problems, be happy and ignore what I write. That is why I am telling the OP to listen himself and compare and buy based on what he likes better.

Thor

 

[pma]

One major issue is that most people designing speaker drivers barely understand how amplifiers functions on anything but the grossest Makro level and most people designing amplifiers understand how speakers functions on anything but the grossest Makro level.

As a result most amplifiers are designed contrary to the requirements of the speakers and visa versa. This is made even worse by an excessive reliance on objective tests that have no demonstrated reliable correlation with "good sound" in both camp's.

That's very true.

 

[pma]

Linear, even zero distortion voltage source creates distorted current if the load impedance is nonlinear. Speaker impedance is nonlinear. So the only way to reduce speaker distortion is a linear current source. Then we get frequency response modulated by speaker impedance. To make it flat, we need EQ. That’s it. It is the only way to reduce speaker intrinsic distortion.

Proudove buzeni repro a zkresleni repro

 

[theREALdotnet]

Linear, even zero distortion voltage source creates distorted current if the load impedance is nonlinear. Speaker impedance is nonlinear. So the only way to reduce speaker distortion is a linear current source. Then we get frequency response modulated by speaker impedance. To make it flat, we need EQ. That’s it. It is the only way to reduce speaker intrinsic distortion.

Non-linear impedance is not the only cause of speaker distortions, so what you’re proposing cannot be the only way of reducing them.

 

[Thorsten Loesch]

Linear, even zero distortion voltage source creates distorted current if the load impedance is nonlinear.

Correct. And the distortion mechanism is a cubic function for both current and frequency. Which brings me back to tweeter voice coils and 100's of mV of Carrier breakthrough, as opposed to 100's of uV, which is what should be done.

Incidentaly, heavy copper cladding and copper caps on the pole piece can reduce the nonlinear voice coil impedance considerably, but they require larger magnets magnetised to higher strength to retain the magnet system efficientcy.

But let's not forget other speaker nonlinearities.

The magnet field in which the voice coil travels is nonlinear. Modern FEM Analysis can be used to create magnet systems that reduce this considerably.

Dynaudio were pioneers of low distortion magnet systems, with their "inside out" magnet systems and very large voice coils.

The mechanical parts of the suspension are subject to nonlinearities and hysteresis.

So more than current mode is needed. Modern driver designers are getting fairly good at playing off non-linearities to create low distortion drivers that operate with voltage drive.

Speaker impedance is nonlinear. So the only way to reduce speaker distortion is a linear current source. Then we get frequency response modulated by speaker impedance. To make it flat, we need EQ. That’s it. It is the only way to reduce speaker intrinsic distortion.

I build active speakers (DIY/Research project) in the 1980's, which used current drive for the MF & HF system (East German 25mm Supranyl Dome & 130mm Schulze KG wideband Studio driver) and current indeed reduced distortion and the other problems were easily equalised.

I failed to make current drive work in a meaningful way for bass until I added motional feedback. As I had dual 200mm Woofers, i was able to use a bridge circuit to derive the back-emf.

Despite the intrinsic nonlinearities of this sensing method it lowered distortion a lot, suggesting that mechanical nonlinearities and not magnetic field nonlinearities were dominant with the drivers.

The motional feedback system did have a tendency to overcompensate when played very loud as the voice coil left the magnet gap, resulting in burned woofers. I intended to fix that with the same kind of excursion and clipping based limiting I had already employed in Sound reinforcement system electronics.

Before I got around to that the political situation in my home country deteriorated considerably and I elected escape to the west rather than waiting around for the 3am knock on the door.

When I eventually was able to return after the velvet revoltions, my apartment had been completely stripped of everything of value, including of course all audio gear, my Volga Car and my Motorcycles were gone as well and nothing elwas ever found again.

The work on that speaker did inform some abortive projects at ifi, not the least the "iMon" active speaker/monitor, which used pressure sensing based motional feedback for the woofer.

Crossover at 1.2kHz, phase coherent type crossover, fully discrete Class A /J-Fet/Mosfet based input, EQ and active crossover, 500W 8R Class D amplifier for Woofer. The elephant trunk in front of it contains the pressure sensor, a modified electret microphone with highly specialized electronics. Discrete Class AB (heavy Class A Bias) with 50W for the HF.

Current drive was not employed, as it did not benefit the drivers used. Woofer is Wavecor with a low distortion magnet system, HF is custom with a low distortion magnet system and very large snail shaped rear chamber.

It was a really good compact speaker and played quite loud and low cleanly. It was cancelled because dealers and distributors claimed that a "speaker from an electronic brand will not sell". Personally I just think they were trying to protect sales of their existing high margin passive speaker lineup by killing of a potentially revolutionary product.

Thor

 

[pma]

But let's not forget other speaker nonlinearities.

Yes, quite well described by Wolfgang Klippel

So more than current mode is needed.

Sure, but as seen in Klippel's images above many distortion mechanisms are improved by current drive.

The motional feedback system did have a tendency to overcompensate when played very loud

I can imagine that.

Current drive was not employed, as it did not benefit the drivers used.

I agree, the drivers have to be tested if they benefit from current drive or not.

 

[DonH56]

See we're already off-topic, is there a simple easy function to use for first-order modeling of speaker distortion? In my cascaded distortion analysis program I used a tanh() function to model circuit nonlinearity (modified by feedback factor), could have used a square-law function for FETs but stuck with old-school bipolar, but did not include anything for speakers as I was not sure what function to use. A tanh() plus signal scaling to emulate feedback works well for differential circuits, and adding a bit of offset was an easy way to add even-order distortion that still followed the curve (instead of x^n functions), but I wasn't sure what to do for a speaker.

 

[Thorsten Loesch]

See we're already off-topic, is there a simple easy function to use for first-order modeling of speaker distortion? In my cascaded distortion analysis program I used a tanh() function to model circuit nonlinearity (modified by feedback factor), could have used a square-law function for FETs but stuck with old-school bipolar, but did not include anything for speakers as I was not sure what function to use. A tanh() plus signal scaling to emulate feedback works well for differential circuits, and adding a bit of offset was an easy way to add even-order distortion that still followed the curve (instead of x^n functions), but I wasn't sure what to do for a speaker.

There is no simple function. Simply look here:

I use a simple gross first order approximation of "if you can see diaphragm movement in a Speaker it is distorting grossly".

Most drivers I measure have dominant H2 above a few 100Hz, below that it's crap shooting in the dark.

Thor