Stereophile | January 2017 Issue

[microstrip]

(...) Typical SET warmth is pretty much 2nd harmonic, let's call a spade a spade. (...)

Just because some SETs have a lot of second harmonic does not imply their typical sound is due to it. If thinks were so simple , we could easily design a push pull with SET sound easily - we would just to add a stage to create the needed second harmonic. Or even emulate it using SS ...

IMHO the real technical reason behind the sound of an amplifier in stereo systems is not well understood. It is a complex sum of many characteristics. Just focusing on harmonic distortion spectra, ignoring the dynamic aspects, will create misleading generalizations.

 

[microstrip]

You thoughts are interesting, I will add my view on the subject.

(...)

My research into this matter (because it is one of those things in audio that I think i worth being concerned about...not frequency response) indicates that for tube amps the problem lies in 4 areas:

1) Output transformer quality. Normally it is too small and saturates...this is to balance with good HF extension sometimes or more likely to save money. As you can see from many online measurements avvailable, many tube amps (like the Dehavilland GM70 and BAT VK60) have a sharp increase in distortion in the bass...particularly as the power goes up. This creates a whole spectrum of harmonics that are most certainly audible and will contaminate the whole frequency range leading to audible coloration.
2) Underperforming driver sections that are overloaded long before the output tube itself is overloaded.
3) Cheap parts quality. In amps with no negative feedback parts quality is paramount for getting good sound (this is a strong Audio Note and Kondo Philosophy but it holds true). Parts cannot fix bad designs but they can reduce the quality of a good design
4) Power supply robustness and cleanliness. This is pretty true for all electronics but I include it here especially because of potential IM distortion issues with power supply modulation.

If it was just these 4 areas thing would be very simple to solve. Anyone can easily build an amplifier fulfilling these requirements and it will not sound SOTA. We also have SOTA designs that only satisfy some of these items. I would love to know the unknown items 5) , 6), 7) ...

BTW, most exotic parts do not have quality in the electrical sense - but yes, they seem to add to sound quality.

SS amps have other issues:
1) Rising distortion with frequency, which can be seen often in measurements of THD vs. Frequency. This also plagues PP tube amps with negative feedback. This makes sound more fatiguing and hurts tone and dimensionality of stereo.
2) Completely out of whack distortion pattern. Dominates on odd and high order harmonics and even though they are low in level they are often audible and add unpleasant artifacts to the sound. One of the main causes of unnatural "synthetic" or fatiguing sound. Also kills soundstage and image 3d and palpability
3) Overdamped bass. Not such a problem if one has underdamped speakers (common) but with critically damped speakers or horns the sound becomes overly tight and lacking texture
4) Susceptibility to back EMF from highly reactive speakers giving IM distortion, which leads again to issues with point 2).

For the record, Quad amps sound nothing like a good SET...that 1 out of 10 was sadly right...no push pull amps do as it is sort of inherent in push/pull to cancel even harmonics...this changes the character.

Interestingly, BAT's white paper on the VK60 indicates that even as little as 3db of negative feedback had a detrimental impact on the sound according to them. Pity they didn't take care of business and put serious iron on the output or they might have had an even better amp. The VK60 and 75 were always good but not great and that might have been why. I had a VAC 30/30 with adjustable feedback (from 0 to 6db) and it ALWAYS sounded better with 0, clicking on even 2db dried out the sound of the mids and highs...not a good thing.

You are now referring to subjective sound qualities according to your sound and speaker preferences. I have listened to SS sounding great with none of the negative sound attributes you refer. Manufacturer whitepapers just reflect their design options, according to their sound preferences.

How much iron is enough? Well, the Aries Cerat Diana (one of the cleanest sounding amps I have heard) has 18Kg output transformers (big part of why this amp is over 100kg total)! The ones in my Ayon are probably over 10kg each and in my lowly JJ they are well over 10Kg each (it measures quite good in this regard...). More common sizes though are 3-8Kg and this is probably too small for anything other than a single 300B. The problem with large transformers is that they have difficulties with high frequencies unless wound very carefully and thoughtfully.

Considering the weight of transformers I will simply ask you a question - how do you scale weight with output power in a power transformer for clean sound? Should we just consider that what is needed is iron mass in absolute terms?

BTW, do you have measurements for the Aries Cerat? As you seem to consider it your reference in sound quality it would be great to look at them.

 

[853guy]

If it was just these 4 areas thing would be very simple to solve. Anyone can easily build an amplifier fulfilling these requirements and it will not sound SOTA. We also have SOTA designs that only satisfy some of these items. I would love to know the unknown items 5) , 6), 7) ...

BTW, most exotic parts do not have quality in the electrical sense - but yes, they seem to add to sound quality.

Hi Micro,

I don’t think 1) - 4) are by any means a definitive list, but they do shed light on some of the more common problems. Certainly, they mirror the issues the designers I’ve talked with have to contend with (and make sense in light of the number of SETs I’ve heard that have not performed well), but even when addressed do not make up for a lack of concept and/or implementation. Some of my most recent experiences have lead me to discover both concepts and implementations that were perviously unknown to me (easy to achieve, actually), and only through the vision and competence of the designer have been able to achieve a level of playback that for me, provided a significant shift in what I thought possible. Tube X, circuit Y and transformer Z are, in and of themselves, woefully inadequate to achieve SOTA levels - hence my personal preference for the designer, not the topology.

 

[microstrip]

Hi Micro,

I don’t think 1) - 4) are by any means a definitive list, but they do shed light on some of the more common problems. Certainly, they mirror the issues the designers I’ve talked with have to contend with (and make sense in light of the number of SETs I’ve heard that have not performed well), but even when addressed do not make up for a lack of concept and/or implementation. Some of my most recent experiences have lead me to discover both concepts and implementations that were perviously unknown to me (easy to achieve, actually), and only through the vision and competence of the designer have been able to achieve a level of playback that for me, provided a significant shift in what I thought possible. Tube X, circuit Y and transformer Z are, in and of themselves, woefully inadequate to achieve SOTA levels - hence my personal preference for the designer, not the topology.

Surely.

I always remember the two low power amplifiers designed by Tim Paravicini in the 80's. Different topologies, one tube, the other was SS and both sounded exactly the same. He is known to have said that he could achieve the same results with both, but he went on designing with tubes because he enjoyed more designing with these devices than with transistors.

 

[853guy]

Surely.

I always remember the two low power amplifiers designed by Tim Paravicini in the 80's. Different topologies, one tube, the other was SS and both sounded exactly the same. He is known to have said that he could achieve the same results with both, but he went on designing with tubes because he enjoyed more designing with these devices than with transistors.

Great example. I had an extended dalliance with the EAR 534 which drove the Living Voice OBX-RW's the best out of any amp I tried. Also had the Musical Fidelity A1 he had a hand in designing (my first ever amp), although I hear he regretted it once it went into production and began catching fire.

 

[morricab]

You thoughts are interesting, I will add my view on the subject.



If it was just these 4 areas thing would be very simple to solve. Anyone can easily build an amplifier fulfilling these requirements and it will not sound SOTA. We also have SOTA designs that only satisfy some of these items. I would love to know the unknown items 5) , 6), 7) ...

BTW, most exotic parts do not have quality in the electrical sense - but yes, they seem to add to sound quality.



You are now referring to subjective sound qualities according to your sound and speaker preferences. I have listened to SS sounding great with none of the negative sound attributes you refer. Manufacturer whitepapers just reflect their design options, according to their sound preferences.


Considering the weight of transformers I will simply ask you a question - how do you scale weight with output power in a power transformer for clean sound? Should we just consider that what is needed is iron mass in absolute terms?

BTW, do you have measurements for the Aries Cerat? As you seem to consider it your reference in sound quality it would be great to look at them.

It was not intended to be exhaustive...just the biggest most obvious errors. I disagree with you though that it is so easily solved...too many examples of at least one of those 4 conditions not being met.

There are probably some dynamic measurements that could be done on parts to distinguish the good from the bad but you are right that static measurements don't reveal too much.

I have not heard one SS amp that sounds truly natural...sorry I have not and I have heard all the top players with the so-called top speakers they are supposed to work with. You can debate this endlessly but this is what I hear both at shows, dealers and in other people's homes. The Spectral guys want to persuade me, the Mola mola guys want to persuade, the darTZeel guys want to persuade and so on but to my ears there is something important not there.

I do not have data on Areis Cerat.

Core size is important for saturation. Big core, low saturation. The metal type matters too. If it has a high nickel content or is amorphous it needs to be bigger for a given amount of current you will flow through it. Metal laminate thickness matters too...generally the thinner the better, especially for low level resolution. Also, type matters, Double "C" need to be bigger for a given application.

However, large cores will have issues at high frequencies unless really sophisticated winding techniques are adopted. If you have measurements you can see right away if the core used was big enough because either the distortion will go up sharply at low frequencies and as power increases or it will not. If yes, the core is saturating and is too small and if not then it is adequately sized. I have seen some expensive amps though with tiny output transformers. A SET that can deliver 30 watts or more should have a rather big output tranny. Diana makes 25 watts and has 18Kg tranny. My 30 watt Ayon probably has 12+ Kg ones. I don't know of another amp with that power with that size transformer. Some old Audio Research D250 a friend of mine had used similar sized output transformers but this is Push/pull, which can be smaller for a given power requirement.

 

[morricab]

Hi Micro,

I don’t think 1) - 4) are by any means a definitive list, but they do shed light on some of the more common problems. Certainly, they mirror the issues the designers I’ve talked with have to contend with (and make sense in light of the number of SETs I’ve heard that have not performed well), but even when addressed do not make up for a lack of concept and/or implementation. Some of my most recent experiences have lead me to discover both concepts and implementations that were perviously unknown to me (easy to achieve, actually), and only through the vision and competence of the designer have been able to achieve a level of playback that for me, provided a significant shift in what I thought possible. Tube X, circuit Y and transformer Z are, in and of themselves, woefully inadequate to achieve SOTA levels - hence my personal preference for the designer, not the topology.

Yeah, I agree that some designers "get it" and know how to complete the package.

 

[KeithR]

I have not heard one SS amp that sounds truly natural...sorry I have not and I have heard all the top players with the so-called top speakers they are supposed to work with. You can debate this endlessly but this is what I hear both at shows, dealers and in other people's homes. The Spectral guys want to persuade me, the Mola mola guys want to persuade, the darTZeel guys want to persuade and so on but to my ears there is something important not there.

I do not have data on Areis Cerat.

Core size is important for saturation. Big core, low saturation. The metal type matters too. If it has a high nickel content or is amorphous it needs to be bigger for a given amount of current you will flow through it. Metal laminate thickness matters too...generally the thinner the better, especially for low level resolution. Also, type matters, Double "C" need to be bigger for a given application.

However, large cores will have issues at high frequencies unless really sophisticated winding techniques are adopted. If you have measurements you can see right away if the core used was big enough because either the distortion will go up sharply at low frequencies and as power increases or it will not. If yes, the core is saturating and is too small and if not then it is adequately sized. I have seen some expensive amps though with tiny output transformers. A SET that can deliver 30 watts or more should have a rather big output tranny. Diana makes 25 watts and has 18Kg tranny. My 30 watt Ayon probably has 12+ Kg ones. I don't know of another amp with that power with that size transformer. Some old Audio Research D250 a friend of mine had used similar sized output transformers but this is Push/pull, which can be smaller for a given power requirement.

So basically your position is now that all SS and PP amps inherently are flawed. MikeL, Fremer, Harley, etc. all have chosen SOTA SS amps over SET. I know there are a few former Lamm owners that have switched to Dart as well. So to say there isn't a single good SS amp to me doesn't make much sense. Listening is a big part of analysis, so can't be ignored as these owners have proven. Even if your technical contention is right, most modern speakers don't work well with SET (at all) and would be subject to frequency response errors and other distortion issues that you don't deem important, but which totally miss the point of running SET...which is to run one at its lowest distortion factor at very low wattage. The "inner detail" that Ralph/Atmasphere discusses. Most of the "high power" SETs mentioned have bigtime distortion woes and are based on tubes that aren't particularly linear at all. See Wavac measurements.

Transformer design isn't all about weight as amplifier speed is important. SET designs require good iron, agreed as they rely on the OPT so much. But I've had massive double C transformers on a 90lb. Tri SET amp that quite frankly was terribly sluggish with poor dynamics. So balance is definitely preferred. PP design doesn't really need 50lb tformers to avoid core saturation. The amps you tend to bring up are 15 year old designs fyi.

Your positions - while aggressive - are still quite subjective as microstrip alludes to despite relying on "studies" that you deem are correct. The reason being (I think) that people don't hear distortion the same.

 

[bonzo75]

So basically your position is now that all SS and PP amps inherently are flawed. MikeL, Fremer, Harley, etc. all have chosen SOTA SS amps over SET. I know there are a few former Lamm owners that have switched to Dart as well. So to say there isn't a single good SS amp to me doesn't make much sense. Listening is a big part of analysis, so can't be ignored as these owners have proven. Even if your technical contention is right, most modern speakers don't work well with SET (at all) and would be subject to frequency response errors and other distortion issues that you don't deem important, but which totally miss the point of running SET...which is to run one at its lowest distortion factor at very low wattage. The "inner detail" that Ralph/Atmasphere discusses. Most of the "high power" SETs mentioned have bigtime distortion woes and are based on tubes that aren't particularly linear at all. See Wavac measurements.

Transformer design isn't all about weight as amplifier speed is important. SET designs require good iron, agreed as they rely on the OPT so much. But I've had massive double C transformers on a 90lb. Tri SET amp that quite frankly was terribly sluggish with poor dynamics. So balance is definitely preferred. PP design doesn't really need 50lb tformers to avoid core saturation. Check out ARC's Ref75 measurements - weighs less than 50lbs and measures quite well. The amps you tend to bring up are 15 year old designs fyi.

Your positions - while aggressive - are still quite subjective as microstrip alludes to despite relying on "studies" that you deem are correct. The reason being (I think) that people don't hear distortion the same.

Yes Greg Baron's client moved from all top of the line NAT to Dartzeel and they both think it better. Gian and friends preferring CH Precision to triodes. Btw, I think this thread should be renamed to Science on, Science off.

 

[morricab]

So basically your position is now that all SS and PP amps inherently are flawed. MikeL, Fremer, Harley, etc. all have chosen SOTA SS amps over SET. I know there are a few former Lamm owners that have switched to Dart as well. So to say there isn't a single good SS amp to me doesn't make much sense. Listening is a big part of analysis, so can't be ignored as these owners have proven. Even if your technical contention is right, most modern speakers don't work well with SET (at all) and would be subject to frequency response errors and other distortion issues that you don't deem important, but which totally miss the point of running SET...which is to run one at its lowest distortion factor at very low wattage. The "inner detail" that Ralph/Atmasphere discusses. Most of the "high power" SETs mentioned have bigtime distortion woes and are based on tubes that aren't particularly linear at all. See Wavac measurements.

Transformer design isn't all about weight as amplifier speed is important. SET designs require good iron, agreed as they rely on the OPT so much. But I've had massive double C transformers on a 90lb. Tri SET amp that quite frankly was terribly sluggish with poor dynamics. So balance is definitely preferred. PP design doesn't really need 50lb tformers to avoid core saturation. The amps you tend to bring up are 15 year old designs fyi.

Your positions - while aggressive - are still quite subjective as microstrip alludes to despite relying on "studies" that you deem are correct. The reason being (I think) that people don't hear distortion the same.

Yes, my position is that PP, Class AB and use of negative feedback is inherently flawed. They generate distortions that are psychoacoustically suboptimal. Using negative feedback to "correct" them might help them sound less offensive (imagine what a SS amp would sound like with 10+% of THD without the feedback loop!) but the patterns are all wrong for natural sound. Maybe someday a perfectly linear amplification device will be invented and we can finally have distortion free sound but until that day i will look for devices that produce harmonically sympathetic patterns that our ear/brain can effectively mask.

Since most SS amps are PP then that makes most of them inherently flawed; however, a SET with transistors is potentially really good sounding. I myself not so long ago had a NAT Symbiosis SE, which was a hybrid SET with a single large MOSFET as the output stage for each channel. It was not as good as some other SETs I have owned but it showed that way has potential I think. I would love to hear the Ypsilon SET100 someday. KR Audio is also a hybrid but in reverse, using FETs for small signal and driver and tube output. We compared head to head NAT Symbiosis SE and KR VA350i and the KR won that showdown. Another technology I would love to try is the SET OTL. Two companies make them: Aries Cerat and Transcendent Sound (as a kit). The Aries Cerat I think is without feedback and the Transcendent sound will have feedback.

What MikeL, Fremer, Harley etc. have chosen is of no consequence to me. If you read the Fremer review of the Lamm ML3 you can see the comments that indicate MF is aware of the tonal nature of the Lamm vs. the more skeletal "speed" approach of the darTZeel. MF is clearly distressed by amplifiers that measure "poorly" (his KR review many years ago showed this...he thought it sounded amazing but felt disgraced by the measurements) and so most likely has an ego issue regarding amp choice. He is just not thinking psychoacoustically though. No idea about MikeL or Harley.

"A strong case could be made for the verisimilitude of the sound from either of these pairs of similarly priced amplifiers, but the award for mesmerizingly hypnotic sound must go to the Lamm ML3 Signature.
Read more at http://www.stereophile.com/content/lamm-industries-ml3-signature-monoblock-power-amplifier-page-2#al5p44QtMBOicYTP.99"

That is not to say though that a SET cannot sound worse than other topologies if designed or executed poorly...there are of course numerous examples of this as well.

Your point about transformers doesn't make sense. You are making an assumption that a large transformer is inherently slower than a small one? This has no meaning. Perhaps the transformers in your Tri SET amp were simply not very well designed. If the laminations are too thick or the winding not done well enough to prevent HF rolloff etc. But it is clear that to eliminate issues with saturation you need big iron...or go OTL and deal with other issues.

 

[microstrip]

Yes, my position is that PP, Class AB and use of negative feedback is inherently flawed. They generate distortions that are psychoacoustically suboptimal. (...)

Can you please tell us quantitatively what is an optimal spectra of distortions according to your views?

 

[morricab]

Can you please tell us quantitatively what is an optimal spectra of distortions according to your views?

it is the pattern that follows the ear/brains own self-generated distortion pattern. As Cheever puts it:

"The ears’ self generated harmonics mask external harmonic distortion that has the
same character. The ears’ harmonic distortion is fully studied and falls off at a rate of
approximately 10^n. I propose that external harmonics strictly adhering to this
envelope are fully “undistorted” by our ear-brain system and are thus
indistinguishable from pure tones."

So, this is in effect an exponential decay from n = 2 to n = infinity. Or in other words monotonic.

"The increase of aural harmonics follow sound pressure level increases non-linearly
and at different rates per harmonic. Therefore absolute system SPL must be
considered."

"Intermodulation distortion is masked by this same mechanism. Amplifier topologies
exist that are free from dynamic intermodulation affects, and whose residual
intermodulation is linearly related to harmonic distortions."

"The character of the noise envelope within a sound transient is important to the brains
recognition system. Noise floor pollution via low level high order I.M. products are to
be avoided."

"Olson from RCA/Victor R&D Labs continued testing the first
8 harmonics and over a broad range of sound pressure levels, reproduced here in Fig. 2-2"

"This has been redrawn for clarity in Fig 2-3. Notice that the ear creates significant levels
of the second harmonic, nearly 10% of the fundamental for sound pressure levels (SPL’s)
of 90dBA and above. Also the slope of the harmonic reduction versus input reduction
varies with the harmonic power, beginning at approximately 1:10 for the 3rd harmonic to
1:1 for the 9th harmonic."


"A reduced SPL range is shown. Even for the moderate S.P.L. of 80dBA, the 2nd
harmonic is at the equivalent of 65dBA or normal voice level, and the 3rd at 45dB. This is
still ~40dB above the mid-band threshold of hearing, yet one does not hear the
harmonics! Only a single pure tone is heard. The ear/brain appears to be able to
completely suppress the sound of a range of harmonics if they conform to this specific
pattern."

"This pattern is the aural harmonic envelope. It follows that this same mechanism
will mask harmonics arising in the sound reproduction chain if they follow this pattern. If
the harmonics do not follow this pattern, the ear brain indeed detects these as new tones.
Therefore, for all but extreme frequencies and sound pressure levels, any electronics that
generate this harmonically consonant envelope will be transparent."




"It is a mathematical expression relating the percentage of the fundamental S.P.L.
of the ears self distortion, per harmonic, relative to the sound field S.P.L.

%Fn = (1.35 *10^(dBA/22))/n^11 = Eq. 2-1 Individual Aural Harmonics

Where: %Fn= Aural Harmonic Amplitude in % of Fundamental for the nth
harmonic.
dBA = Decibels “A” weighted Sound Pressure Level resultant from the
Fundamental.
n = The harmonic number. f = nFf where f is frequency,
Ff= fundamental frequency


"The power of the exponentiation may seem high but the fit is excellent, shown
following in Fig. 2-7."

"In calculating the magnitude of an amplifiers deviation for the aural harmonic envelope I
propose that each harmonics deviation be on a relative basis (% of reading, referenced to
the level of the nth aural harmonic derived from Eq. 2-1), rather than on the absolute
percentage referenced to the fundamentals level"

Based on all my reading, this approach makes the most overall sense as it takes into account what is known about ear/brain self-generated harmonics, masking and SPL.

This two are also going in a similar direction.
http://www.gedlee.com/Papers/Distortion_AES_I.pdf
http://www.gedlee.com/Papers/Distortion_AES_II.pdf


Based on looking at tons of measurements and also theory, one can see that push/pull can never generate this kind of distortion pattern, unless the designer deliberately finds a way to inject even order distortion into the circuit to counteract the even order cancellation that is inherent in the push/pull concept. Only single ended does this pattern naturally. Whether you agree that the theory about the masking and aural harmonics by Cheever is valid or not is another question. However, Geddes found similar things. What cannot be disputed is that the ear/brain does in fact make a monotonic pattern of its own with every sound that makes it vibrate and the brain dismisses this or at least that is what is defined as "pure". Masking is well known phenonmenon. Cheever is claiming that the signal "hides" in the self-generated pattern if the device mimics the pattern. The tighter it mimics the more pure it will sound.

 

[PeterA]

Since most SS amps are PP then that makes most of them inherently flawed;

Morricab, Have you heard single ended Class A SS amplifiers, such as some of the Pass Labs designs? If so, did they exhibit the same kinds of distortions typical of push-pull SS designs? I don't know enough about the technology, but my understanding is that their sonic signatures, distortion patterns, are somewhat different, and more in line with what the ear/brain hears in nature.

 

[morricab]

Morricab, Have you heard single ended Class A SS amplifiers, such as some of the Pass Labs designs? If so, did they exhibit the same kinds of distortions typical of push-pull SS designs? I don't know enough about the technology, but my understanding is that their sonic signatures, distortion patterns, are somewhat different, and more in line with what the ear/brain hears in nature.

it won't be quite the same based on my experiences with my NAT Symbiosis. But then the transfer function of a MOSFET is different from a triode so that character is there. Also, a SE(transistor) doesn't have an output transformer, which also affects the character. The Symbiosis I had was quite promising in some ways and vaguely dissatisfying in others. It was hard to put my finger on it.

I would love to try a pure SS SE(T) like a Pass Firstwatt or another hybrid like the Ypsilon SET100s. I would also like to try a SET OTL tube amp.

There is a paper from Boyk and Sussmann where they modelled how all these devices behave singled ended, push/pull and differential stages...all Class A. Class AB will make much worse distortion.

https://www.its.caltech.edu/~musiclab/feedback-paper-acrobat.pdf

Interesting stuff.

 

[ack]

There is a paper from Boyk and Sussmann where they modelled how all these devices behave singled ended, push/pull and differential stages...all Class A. Class AB will make much worse distortion.

https://www.its.caltech.edu/~musiclab/feedback-paper-acrobat.pdf

Interesting stuff.

I read the first few pages, very interesting paper, will have to read the rest when I get time. Thanks for that.

 

[microstrip]

it is the pattern that follows the ear/brains own self-generated distortion pattern. As Cheever puts it:

"The ears’ self generated harmonics mask external harmonic distortion that has the
same character. The ears’ harmonic distortion is fully studied and falls off at a rate of
approximately 10^n. I propose that external harmonics strictly adhering to this
envelope are fully “undistorted” by our ear-brain system and are thus
indistinguishable from pure tones."

So, this is in effect an exponential decay from n = 2 to n = infinity. Or in other words monotonic.

(...)

Based on looking at tons of measurements and also theory, one can see that push/pull can never generate this kind of distortion pattern, unless the designer deliberately finds a way to inject even order distortion into the circuit to counteract the even order cancellation that is inherent in the push/pull concept. Only single ended does this pattern naturally. Whether you agree that the theory about the masking and aural harmonics by Cheever is valid or not is another question. However, Geddes found similar things. What cannot be disputed is that the ear/brain does in fact make a monotonic pattern of its own with every sound that makes it vibrate and the brain dismisses this or at least that is what is defined as "pure". Masking is well known phenonmenon. Cheever is claiming that the signal "hides" in the self-generated pattern if the device mimics the pattern. The tighter it mimics the more pure it will sound.

Suppose we accept the monotonic distortion decrease theory. Why does this pattern need to be natural (intrinsic) of the output stage? Why can we add it at the SE output and not at the input - an input stage with this pattern of distortion and other stages, including power with minimal distortion? How does the ear/brain know where the distortion is produced?

Do you consider feedback a nuisance because it changes this pattern of distortion or by another specific technical reason?

BTW, Mike Lavigne should be smiling - the DartZeel 458 is one of the few solid state amplifiers that approaches the monotonic distortion pattern!

 

[bonzo75]

BTW, Mike Lavigne should be smiling - the DartZeel 458 is one of the few solid state amplifiers that approaches the monotonic distortion pattern!

Which are the others?

 

[microstrip]

I read the first few pages, very interesting paper, will have to read the rest when I get time. Thanks for that.

I have read it sometime ago - this article is considered a classical and is really worth reading. It is based on simulations of simple small signal models - no measurements of real devices at all. However it is no way conclusive (quoting from the conclusion) "Similarly, the fact that feedback can sometimes increase the relative distortion of very low-level signals makes it tempting to relate this finding to the loss of fine detail and room sounds. Whether or not these are appropriate attributions can be determined only by psychoacoustic experiments."


James Boyk is president, A&R director, musical director, recording engineer, production manager, jacket notes author, and the star performer. He is also Artist in Residence and lecturer in music and engineering (specifically sound recording and reproduction) at Cal Tech. And he happens to be one of digital's most ardent detractors, having conducted, and widely publicized, several controlled listening tests that proved to his satisfaction that digital recordings are destructive to musical sound. (I will not question his methodology or conclusions here; suffice it to say that James is as stalwartly pro-analog and anti-digital as it is possible to be.) (from the Stereophile site)

And yes, I own and appreciate his excellent sounding All Tube Analog LP of piano works!

 

[microstrip]

Which are the others?

There is another well know one, still trying to remember ... I only memorized the more more expensive ...

 

[KeithR]

it is the pattern that follows the ear/brains own self-generated distortion pattern. As Cheever puts it:

"The ears’ self generated harmonics mask external harmonic distortion that has the
same character. The ears’ harmonic distortion is fully studied and falls off at a rate of
approximately 10^n. I propose that external harmonics strictly adhering to this
envelope are fully “undistorted” by our ear-brain system and are thus
indistinguishable from pure tones."

So, this is in effect an exponential decay from n = 2 to n = infinity. Or in other words monotonic.

Based on looking at tons of measurements and also theory, one can see that push/pull can never generate this kind of distortion pattern, unless the designer deliberately finds a way to inject even order distortion into the circuit to counteract the even order cancellation that is inherent in the push/pull concept. Only single ended does this pattern naturally. Whether you agree that the theory about the masking and aural harmonics by Cheever is valid or not is another question. However, Geddes found similar things. What cannot be disputed is that the ear/brain does in fact make a monotonic pattern of its own with every sound that makes it vibrate and the brain dismisses this or at least that is what is defined as "pure". Masking is well known phenonmenon. Cheever is claiming that the signal "hides" in the self-generated pattern if the device mimics the pattern. The tighter it mimics the more pure it will sound.

Haven't we already debunked Cheever? I mean come on Brad. It's a dude who wrote a thesis based on a 5 person test on 2 crappy amps 20 years ago. It was never reviewed or presented to AES or anything. The guy isn't even in the audio world today, but a consultant. Yet you ignore Nelson Pass who says a third of people prefer 3rd harmonic and has 100x over the experience of Cheever.

Ironically you haven't posted measurements of the Aries Cerat which you are so in love with now. I'm going to die laughing if it has a bit of NFB lol.