Modulus-86: Composite amplifier achieving <0.0004 % THD+N.

[maty tinman]

If you use the search engines will find that more people complain about it (fatigue, tiring...).

It is assumed that quality music listening and very well recorded. For the current commercial music (so badly recorded) will be enough class D.

Maybe the new implementations (class D + class A hybrids) of Devialet (very expensive) and Nuprime have solved that problem, maybe.

-> Nuprime/Nuforce vs Devialet http://www.audiocircle.com/index.php?topic=135779.0\

ICEpower, Hypex and others do not resolved it.

For me, the most problematic of using LM3886/LM4780 is its low 14 microV/S. Maybe with orchestral works (I listen a lot) is insufficient, maybe. As there are no user reviews of the amplifiers designed by Tom I still in doubt.

 

[soongsc]

I think it is not a good idea to categorise Class D as having a specific quality. Different people just have different experiences. Just a few days ago, a team I was giving a presentation to had one engineer identify Class D as sounding flat and without punch. I shared my experience about how the transformer in a linear supply influenced the punch of a Class D. The real issue is with each specific design. It could be the layout, it could be the design, it could be the match between power supply and the amp, it could also be input interface compatibility. Each require detail investigation. The thing is, having good audition feedback can help zero in on the more critical issues faster than using a matrix of tests to zero in on the issue. Then when you know where the problem might be, some changes will help you verify whether sound changes obviously enough to explore the issue further, this is where the proper measurements will help you fine tune a design.

Two Class D amplifiers I do not recall being fatiguing are the TACT Millennium and the New Class D.

 

[Julf]

If you use the search engines will find that more people complain about it (fatigue, tiring...).

If you use search engines, you will find that people complain about pretty much everything possible. That kind of stuff is usually referred to as "anecdotal".

It is assumed that quality music listening and very well recorded

. For the current commercial music (so badly recorded) will be enough class D.

I guess that must be because the widespread use of class D amps in studios these days...

Maybe the new implementations (class D + class A hybrids) of Devialet (very expensive) and Nuprime have solved that problem, maybe.

-> Nuprime/Nuforce vs Devialet Nuprime/Nuforce vs Devialet

ICEpower, Hypex and others do not resolved it.

I guess we have to take your word for it.

For me, the most problematic of using LM3886/LM4780 is its low 14 microV/S. Maybe with orchestral works (I listen a lot) is insufficient, maybe. As there are no user reviews of the amplifiers designed by Tom I still in doubt.

So you mainly base your decisions and opinions on subjective, anecdotal indications rather than measurements or scientific models?

 

[maty tinman]

If you are used over the sound of a valve SET amplifier, the Hypex nc400 become unbearable as he confessed one that sold on eBay.

I pay very close attention to the specifications and measurements but I know we can not trust them at all, the end is the ear, our musical tastes and (very important) the room.

My tests have focused on improving the dismal mains I suffer at home (very RFI/EMI noise, DC for tons and high voltage), getting incredible today for the money spent on results afford it. Spend a fortune on a new equipment without having solved the above made no sense. For now my only problem is the high voltage, near 240V before 20pm (I have thought the solution, after the summer).

You sill also find many who are unable to differentiate MP3 of FLAC and still less that value FLAC 24/96. And even fewer appreciate the excellent vinyl rips I hear.

Yesterday I noticed that the differences between MP3 320kpbs and FLAC 16/44 have increased a lot in this CD rip:

Michel Camilo & Tomatito - Spain Again (2006)

-> http://www.amazon.com/Spain-Again-Michel-Camilo-Tomatito/dp/B000F8ZN7W

--------- --------- ---------- ------------

My second system:

PC --> ODAC --> AV Marantz SR4500 (with cheap mods, in and out) --> KEF Q100 (bass-reflex closed)

Schaffner 3A + Würth ferrite 150Khz 12.5mm --> Power strip (alluminum) --> DC blocker + DC blocker x2 --> Schaffner 3A --> AV Marantz SR4500

Schaffner 3A + Würth ferrite 150Khz 12.5mm --> Power strip (alluminum) --> DC blocker + DC blocker x2 --> Würth ferrite 150Khz 8mm x2 --> Schaffner 3A --> PC (Bequiet! L8 400W)

PC USB --> Würth ferrite 150Khz (two rounds) --> ODAC

 

[billshurv]

amplifier with better measurements is preferred. Shock horror, Toole was right. Move on nothing to see.

 

[Julf]

If you are used over the sound of a valve SET amplifier, the Hypex nc400 become unbearable as he confessed one that sold on eBay.

You are probably right, but I think that says more about the valve SET amp than about the nc400.

I pay very close attention to the specifications and measurements but I know we can not trust them at all, the end is the ear, our musical tastes and (very important) the room.

But then all we can and should say is "I like X, and don't like Y" instead of saying "X is better than Y".

My tests have focused on improving the dismal mains I suffer at home (very RFI/EMI noise, DC for tons and high voltage), getting incredible today for the money spent on results afford it.

I suggest a $15 X10 mains noise filter - deals with most RFI. Where does your RFI/EMI come from?

For now my only problem is the high voltage, near 240V before 20pm (I have thought the solution, after the summer).

Here is how my mains looks like:

What problems does the 240V cause? A constant (slight) overvoltage won't affect sound quality - it just makes things run slightly warmer.

You sill also find many who are unable to differentiate MP3 of FLAC and still less that value FLAC 24/96.

Indeed, considering 24/96 doesn't actually offer anything as a distribution/storage format compared to 16/48 (intermediate processing and recording is another matter). But that is yet another discussion.

Yesterday I noticed that the differences between MP3 320kpbs and FLAC 16/44 have increased a lot in this CD rip:

Michel Camilo & Tomatito - Spain Again (2006)

-> Amazon.com: Michel Camilo & Tomatito: Spain Again: Music

And do you know they come from the same mastering/EQ? Did you check that they are level-matched (as we know, even a 1 dB level difference makes you think the louder one is "better").

 

[maty tinman]

There are things that do not appear in the measurements. Measurements and specifications serve to weed out bad equipments. For example, I prefer an amplifier with S / N 105dB as my Marantz than an expensive audiophile with 90 dB or less, especially since the S/N 102 dB of my ODAC with RCA outputs (with cheap mod inside too).

The drastic reduction of all types of noise (electrical and electronic) entails a substantial improvement in sound quality, as I could verify (only the electrical).

The few times I have had 232.5V - 233.5V the sound was much better. Prior to 238V unbearable had to turn it off, now with 237V sounds good but without the magic of the closest voltage to 230V. Marantz neglected filtering in this amplifier.

I strongly suspect that it was designed to work at 110V/220V and not 115V/230V, with a margin of variation of 7% instead of the usual 10%, but only suspicion.

The DR was the same I think. FLAC rip sounds with minor volume that MP3. I think because I deleted the MP3 rip. I remember I save a copy in HD USB3!

 

[maty tinman]

Repeat.

 

[Julf]

This is getting seriously off-topic, but...

The drastic reduction of all types of noise (electrical and electronic) entails a substantial improvement in sound quality, as I could verify (only the electrical).

So how did you verify it? What aspects of sound quality were improved?

The few times I have had 232.5V - 233.5V the sound was much better

. Prior to 238V unbearable had to turn it off, now with 237V sounds good but without the magic of the closest voltage to 230V. Marantz neglected filtering in this amplifier.

Filtering won't change the actual voltage. It seems you are confusing voltage levels and noise. In what way do you think the amp is being affected by the higher voltage?

The DR was the same I think. FLAC rip sounds with minor volume that MP3. I think because I deleted the MP3 rip.

DR doesn't tell you anything about absolute levels - and small (1-2 dB) absolute volume differences are not really perceived as volume differences, but affect your perception of "sounds better".

 

[maty tinman]

Noise filtered, voltage regulation... in the AV Marantz SR4500 are not good.

I have been solved bottlenecks of my system. The last thing I have to solve is the high voltage. With just over 200 € I believe that I will solve!

Or a new amplifier with very good voltage regulation like Tom's amplifiers or the Akitika GT-101 of Dan Joffe.

-> GT-101 Stereo Power Amplifier

Class AB with a Regulated

The GT-101 has an electronically regulated power supply. That makes it unusual among audio power amplifers. But in some ways, it offers an advantage similar to class A operation.

Class A amplifiers can be among the most linear of amplifiers, and the least efficient. They draw a constant current from the power supply, no matter what signal is being delivered to the speaker. That says that a Class A amplifier doesn't modulate the supply rails..constant current drain produces a constant load on the power supply, and hence a constant voltage.

Is a Class AB amplifier with an electronically regulated power supply almost as good in that respect? If the power supply's output impedance is low, then the output signal minimally modulates the rails. That should cause less distortion at the output. In that sense, the Class AB with regulated power supply should be about as blameless as a Class A amp.

The one way that the Class A amplifier still has it over the Class AB is that at low listening levels, the output device has nearly constant current. That is, its operating current is the large class A bias current with the small signal modulations. Of course, at high output levels, the Class A advantage diminishes, as the output device then experiences wide swings in operating current.

but they only have 14 microV/s. And I would spend more money!

 

[Julf]

Noise filtered, voltage regulation... in the AV Marantz SR4500 are not good.

That might be, but they are two different things, and you haven't told us how the overvoltage is supposed to affect sound quality.

I have been solved bottlenecks of my system. The last thing I have to solve is the high voltage. With just over 200 € I believe that I will solve!

I would start by verifying it actually needs to be solved. Be aware that many ways of regulating voltage actually restrict peak current.

but they only have 14 microV/s. And I would spend more money!

And what, in your view, would be a good slew rate (V/s)? Based on what parameters?

 

[maty tinman]

Orchestral music is more exigent to ALL system. More instruments and very fast "transitions" like silence to all orchestral instruments in the next second.

I think that with 40 V/uS - 80 V/uS or more this sound will be better. Minimum/optimum is 20 V/uS I read, putting in doubt appreciate greater value (NwAvGuy blog posts and other sites I think).

-> http://nwavguy.blogspot.com

commercial music "chunda chunda" wil be not problem with 14.

By the way, the instruments widely appreciated by far the lowest electrical noise are: organ, piano and guitar.

 

[maty tinman]

https://duckduckgo.com/?q=nwavguy%2Bslew%2Brate&t=ffab

All must read this great article:

-> http://nwavguy.blogspot.com/2011/08/op-amp-measurements.html

Slew Rate – Here’s another huge myth: The faster the slew rate the better. But the fastest “slew” you can get from a 16/44 digital recording has a period of about 22 uS. So the O2, worst case, needs to slew 20 volts in 22 uS or a slew rate of 0.9 V/uS. And even considering those rare 24/96 recordings, the requirement is still 1.8 V/uS. SACD or 24/192, in theory, could require about 3.6 V/uS but, in practice, never will as that would be a 0 dBFS signal close to 100 Khz. Such a signal would never make it through the recording chain, and if it did, would fry tweeters and cause bats to fly into the side of your house. The peak levels in music generally fall as the frequency rises. This is especially true at ultrasonic frequencies. If you think I’m wrong, please reference a recording than can challenge even a 1.5 V/uS headphone amp. The industry rule of thumb, even on the recording side of the signal chain, is 0.2 V/uS per volt RMS of maximum output. I’ve doubled that just to be ultra conservative so 7*0.4 = 2.8 V/uS for the O2’s requirement. As I have explained elsewhere,slew rates well in excess of what’s needed come at a price. Somewhere in the range of 2.5 – 5 V/uS is ideal for a high output headphone amp like the O2.

 

[maty tinman]

Maybe it was 20 V/uS to an amplifier 100W at 8 ohms the optimal/minim is what I read.

Somebody bring more light about.

 

[Julf]

Orchestral music is more exigent to ALL system. More instruments and very fast "transitions" like silence to all orchestral instruments in the next second.

So now you are throwing dynamic range into the mix of unrelated things?

I think that with 40 microV/s - 80 microV/s or more this sound will be better.

At what signal level?

I hope you realize that a plain slew rate, without reference to the actual voltage swing, is totally meaningless.

Minimum/optimum is 20 microV/s I read, putting in doubt appreciate greater value (NwAvGuy blog posts).

Where in that link is 20 mV/s mentioned? I hope you also realize that that number is off from the "industry rule of thumb" of 0.2 V/uS per volt RMS of maximum output that you quote by a factor of 10000000?

20 mV/s = 0.02 V/s
0.2 V/uS = 200 000 V/s

By the way, the instruments widely appreciated by far the lowest electrical noise are: organ, piano and guitar.

And what has that to do with... well... anything?

[edit: OK, I see you edited your original postings to fix the unit confusion]

 

[Julf]

Maybe it was 20 V/uS to an amplifier 100W at 8 ohms the optimal/minim is what I read.

100W at 8 ohm would be 28 V, so the "industry rule of thumb" value would be 6 V/us - for the power amp, but not for an op-amp at the input.

 

[maty tinman]

-> Slew Rate in Audio Amplifiers - What Does it Mean? | Audioholics

At this point, an observer may note that the number calculated above is a lot lower than the specifications you are likely to encounter, even on entry level receivers. As an example, the Harman Kardon AVR 1565 boasts a slew rate of 40V/usec. Considering its rated power of 70W per channel into 8 ohms (the above example of 50V translates into a 312.5 watt peak at 8 ohms), 40V/microsecond would appear to be gross overkill. Realistically, it is a matter of good engineering. The bare minimum that we’re calculating is just that; most good linear amplifiers aim for flat response well beyond 20kHz, not because we can hear that high or because we have loudspeakers capable of reproducing such frequencies effectively, but to ensure slew rate related problems are a total non-issue within the audible band. Simply put, once you start tossing in other requirements, i.e. expecting practically zero distortion even at full power into a difficult load at 20kHz (a problem area for some amplifiers, notably Class D), a bit of extra margin in terms of slew rate can come in quite handy.

 

[Julf]

As an example, the Harman Kardon AVR 1565 boasts a slew rate of 40V/usec. Considering its rated power of 70W per channel into 8 ohms (the above example of 50V translates into a 312.5 watt peak at 8 ohms), 40V/microsecond would appear to be gross overkill.

Yes, it is important that an amp has sufficient slew rate to reproduce the highest frequency in the input signal at the required amplitude. But as I keep pointing out, that depends on the amplitude. There is no universal "sufficient slew rate" without specifying the amplitude (voltage swing) it applies to. Thus the critical slew rate for a power amp is very different from an opamp in the input stages.

 

[maty tinman]

-> Slew Rate Part 2

There was a serious point, which was to demonstrate that amplifiers with very high specified slew rates are not necessarily either 'faster' or more accurately amplify transients than those with a lower specification.

In the next example the green step is from an amplifier with slew limit only 20V/µsec, so having no safety margin at all, compared to the red step produced by an amplifier capable of 100V/µsec, 5 times higher. The higher limit is evidently not helpful, the result is worse. The green step amplifier however has gain-bandwidth product 5 times higher, and clearly this is enough to more than make up for the lower slew limit.

​

The usual slew rate specification fails to mention the corresponding output error. If we specify power output or peak output voltage, we would perhaps mention that it is the value at which distortion reaches 1%, so is there any good reason not to use a similar error level for specifying slew rate?

As demonstrated earlier, the accuracy of the output has little or no relationship to the slew rate limit, it is more closely related to amplifier gain-bandwidth product, and is also strongly affected by any low-pass filtering included in the amplifier, so in practice the 1% error level is not very meaningful. We can however deduce something from the input stage design. If the input stage needs a differential input of 1V to drive its maximum current into the compensation capacitor to achieve for example 100V/usec at the output, then for an amplifier gain of 20 and an input step of 1V that slew rate can only be achieved when the step reaches 1V and with a zero output, otherwise the feedback will reduce the input stage differential voltage, and so we must have a 100% output error. If we only want a 1% output error then the output will be 19.8V when the input step reaches 1V, and the input stage differential input will then be just 10mV. It will then be able to drive only 1% of its maximum current into the compensation capacitor and so can reach only 1V/usec, so the specification of slew rate at 1% output error becomes just 1V/usec. This figure will not be improved either by increasing input stage current or by increased degeneration resistors combined with reduced compensation capacitor, but these steps may have other benefits such as reduced input stage distortion.

 

[Julf]

There was a serious point, which was to demonstrate that amplifiers with very high specified slew rates are not necessarily either 'faster' or more accurately amplify transients than those with a lower specification.

So you agree with my point?