Non-feedback amplification topology and frequency


Is it true that non-feedback topology amplification is prone to frequency limitations if not matched with the absolute perfect speaker system?
I'm trying grasp how to get a flat sound from my very much loved new PL5. I have resorted to EQ, that never mentioned term on these forum, to bring back the ultra-highs of cymbals and bells etc. I've also rolled the pre-tubes and switched out my ICs. Still very much a frown freq graph.
I'm willing to get new speakers to help this amp deliver on its promises. Focal seem nicely high in sensitivity.
jmacinnis
Hi Jmacinnis, I'm glad you are happy with your system. My thread that you pointed out really has nothing to do with what makes us happy about our system or about our sound preference. The crux of my post is about whether a the output from a NFB amp is more faithful to the original signal than that of a ZNFB amp. My post has hit a lot of nerves because it seems that some took my post as an attack on their taste in music and equipment, which it wasn't at all. I was initially perplexed by some of the responses, which had nothing to do with my question. Some didn't even bother to read the important article I linked.
I won't disagree that there may be incentive in the high end "industry" to not make things easy for buyers.

There are still multiple recipes towards good soup that an informed buyer who cares enough can follow to acieve their goals. No one recipe has all the advantage!

Luckily, most of the rest of the world gets by fine with mass produced stuff that works well enough to satisfy the average listener.

Really good things never come easy.....
Mapman, I'm pretty sure that part of the reason this is so has to do with that article I link all the time:

http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

If you happen to compare the two amps on a speaker that is in one camp, you are going to come away with a mistaken impression unless you are aware of the different design paradigms.

This is something the industry won't talk about. Its too inconvenient! And its why we have that darn objectivist/subjective debate. I am of the opinion that if the industry was transparent about this stuff it would be a hell of a lot easier to set up a system and a lot less money would go down the loo. But, OTOH the industry thrives on that, so maybe I'm trying to kill the goose that lays the golden egg huh :)
Well, at least paper should be able to disqualify the big offenders. THen it is left up to our ears, which I agree with 100%. And and over the years, my ears tell me there is more to it than meets the eye here.... ie it is NOT so black and white. That is definitely the case if it cannot be measured. A "leap of faith" is required to reach the conclusion in lieu of supporting data. So goes high end audio.....
It really is that black and white. Sort of like being pregnant.

What is poorly understood is our ears are so sensitive to odd orders that we can hear things that cannot be measured. So it might look 'OK' on paper and our ears say something else.

And there is nothing we can do about it. Our ears are the product of millions of years of evolution. Its not going to change just to suit our inadequacies of design.
"The reason I made that comment is that an amplifier that violates one of the most fundamental rules of human hearing will never sound like real music."

Is it really so black and white in practice though which designs truly "violate the rule" and that use of NF means a violation in all cases?

I would think it more shades of grey with the devil being in the details: the relative magnitude of the violation (which varies and should be measurable) and at what point the "fundamental rule" is "violated".
Mapman, they do, but even in high end audio we have to deal with facts occasionally :)

The reason I made that comment is that an amplifier that violates one of the most fundamental rules of human hearing will never sound like real music. Its not so much the failing of the speaker as it is the amp.
"Ultimately you have to be more careful of your speaker choice; this is doubly important because a speaker that requires the amp to have feedback will never be able to sound like real music- it will at best sound like a good hifi."

Absolute statements like this are bound to hit some nerves. That's what helps create discussion, which is usually where a more complete story might become apparent.
Dracule I see this feedback issue hit a nerve as you started a whole new thread about essentially the same thing as this one. I'm quite ok now with Atmasphere's description of the facts and history of this topic. It is up to debate only whether or not you subjectively like the sound of what comes from feeback or not, and isn't that just taste in the end? I prefer vinyl because of the way my ear reacts...it sure isn't the highest quality choice in terms of specs. My original question was just, Do you need to be more carefull when selecting speakers with non feedback amps. Answered.
Today I am the happy owner of a beautiful pair of Cantons that are fully engaging and blow my old system away. It would be fun to play with the feedback knob on your amp though :)
Ralph,

That was a VERY cool video of the Fisher Z-Matic. I'm constantly amazed of what was done...back-in-the-day!

Regards,
Sam
Thanks atmasphere. That is a plausible explanation why SET amplifiers which almost always employ zero NFB sound more involving (dynamic) than reality. I've found SET amps with zero NFB to be less accurate than PP amps using NFB.
Onhwy61, No. Its simply the way it is. One example (if you read the article) is using transistors on ESLs. Its difficult for them to play bass, and often too bright.

Mapman, Correct- never is a long time. The problem is, we are not likely to change the way our ears work. We can do a lot of things in the medical world, but changing the system that our ears employ is not one of them.

Dracule1, I have no worries about criticism. If I had made the stuff up, that might be different. But I can point you to examples easily enough that will show that my article is not something new, but in fact something old:

Here is a link to a Google search: http://www.google.com/search?client=ubuntu&channel=fs&q=fisher+A-80+amplifier&ie=utf-8&oe=utf-8

The very first hit is a YouTube video image of a Fisher amplifier from the 1950s. The control is inscribed "Constant Voltage" at fully counterclockwise, "Constant Power" at noon and "Constant Current" fully clockwise. The control is a feedback control and balances voltage feedback against current feedback. At noon they cancel each other out (zero feedback). So to disprove me you have to disprove history. All my article does is collect the facts in a single place.

Marqmike, unfortunately all amplifiers make odd ordered harmonics. It is the 5th, 7th and 9th that we are concerned about (the 3rd is considered musical to the human ear). When you add negative feedback to any amplifier design, the result will always be an enhancement of odd orders, even if the feedback is successful in removing most of the open loop distortions ('open loop' means the amp operating without feedback).

The reason this is so has to do with the fixed amount of time it takes a signal to move through a circuit. This time is called 'propagation delay'. The issue is that a low frequencies feedback works pretty good, but as frequency is increased, the propagation delay means that the feedback signal is arriving later and later to correct the circuit. In fact a designer has to be careful with feedback because this effect means that at some frequency the feedback as actually positive rather than negative. This is entirely due to the propagation delay in the amp.

Further complicating things is the fact that the negative feedback analysis performed on amplifiers is done with steady-state signals like sine waves. Audiophile OTOH listen to chaotic, constantly changing waveforms arising from real music. If you think that the amp will behave the same with both signals you are incorrect- the amp will be more distorted with constantly changing waveforms than it will be with a sine wave!

If you analyze and amplifier using Chaos Theory, what we see is that the amp with feedback is considered a chaotic system. In fact the formula for feedback in an amplifier is the same as it is for a classic Chaotic System. However, an amp without feedback, thus analyzed does not behave according to the same formulae! This is a major difference between the two.

Interestingly enough, most of this has been known since the 1950s. Anyone interested should read Norman Crowhurst. He wrote a very interesting booklet on amplifiers and negative feedback. It is available as a free download off of the Pete Millet website. His writings apply as much now as they did 50 years ago.

The bottom line here is that amps without feedback, if otherwise designed to be linear, will have less odd-ordered harmonic content if also operated within their linear region. However, this may not mean what most people think so I will give a classic example:

We all know that SETs have a reputation for 'inner detail' and also that they are amazingly dynamic for their small power output. The reason this is so has to do with the way the SET amplifier makes distortion. At low power it essentially makes no distortion at all- this is the source of that 'inner detail'. But at powers above about 1/4 full power, the odd orders begin to make their appearance (masked by the 2nd, 3rd and 4th harmonics). The ear/brain system uses these harmonics to figure out how loud a sound is and since music has a lot of transients, the odd orders start to show up on the transient first. This means the loudness cues are on the transients of the music- presto! the amp *sounds* 'dynamic'.

The fact of the matter is that as a result, in about 90% of all audiophile conversation, the word 'dynamics' can be safely replaced by the word 'distortion' without changing the meaning of the conversation.

Sorry for the long post...
Marqmike, a zero negative feedback amp does produce odd order harmonics. Please read the link I provided. It will answer most of your questions.
Atmasphere, is it so that a non negative feedback design does not have any odd order harmonics? If that is not true where does that leave us? I highly respect your dialog on this forum and I don't want to stop it so my question is a sincere one. I wonder if we are making too big of a deal with NFB in high quality amplifiers. If the overall design of the amp needs minimal NFB how does that in real conditions compare to a non NFB design. Isn't that a competent design then no matter what emitter device is used? Then I don't think we are talking about bright and harsh as noticeable differences are we? Arn't we really talking about competent designs and non competent designs then? Thanks again Atmasphere for your dialog.
Atmasphere, I'm not trying to be disrespectful, but the link you provided written by you is not convincing.
Atmasphere, as Mapman has indicated, you should be prepared to be criticized if you state opinions in absolutes. What you state are facts are contested by others. See below.

http://sound.westhost.com/articles/distortion+fb.htm
"Ultimately you have to be more careful of your speaker choice; this is doubly important because a speaker that requires the amp to have feedback will never be able to sound like real music- it will at best sound like a good hifi."

Hmmm, never is a long long time!

I think that's arguable at a minimum.
And its also true that if you use an amp with no feedback (Power Paradigm) on a speaker that needs it (Voltage Paradigm) you will get a tonal coloration of the types you describe.
Isn't this an awful big caveat?
More often than not, I've found tube amps that use feedback sound more realistic and are more compatible with wider range of speakers than amps that use no feedback. I've looked into why audiophiles think no feedback is sonically superior, but I can't find any good reason to support this contention. Yes, NFB does increase odd order harmonics but the levels are low, and I doubt if anyone can actually hear them reliably. I've found most tube amps with no feedback to have less controlled bass, less dynamics, and subtle softening of the highs.

I did not state an opinion, but fact. Nor is it something I made up- this has been with us for a long time. You might want to read the article that I linked.

It is quite true that there are more speakers designed for amps with feedback than those without. And its also true that if you use an amp with no feedback (Power Paradigm) on a speaker that needs it (Voltage Paradigm) you will get a tonal coloration of the types you describe.

You are incorrect in stating
the NFB does increase odd order harmonics but the levels are low, and I doubt if anyone can actually hear them reliably
The fact of the matter is that slight increase in odd-orders are described as 'bright' harsh' 'hard' 'clinical' etc. by audiophiles, yet we are talking about distortion levels that may only be increased by a few 100ths of a percent! This is because the ear/brain system uses odd ordered harmonics to sort out how loud a sound is.

This is why two amps can measure flat on the bench, but one might sound bright and the other might not.

IOW the ear is considerably more sensitive to these harmonics than it is to human vocal frequencies.

General Electric proved this back in the 1960s- its not anything new.

Again I recommend you read the article and most of this will be made clear.
"Ultimately you have to be more careful of your speaker choice; this is doubly important because a speaker that requires the amp to have feedback will never be able to sound like real music- it will at best sound like a good hifi."

Atmasphere, I have to disagree with your opinion. I've owned both Atmasphere amp (M60) which uses minimal feedback and tube amps that use up to 20 dB of feedback. More often than not, I've found tube amps that use feedback sound more realistic and are more compatible with wider range of speakers than amps that use no feedback. I've looked into why audiophiles think no feedback is sonically superior, but I can't find any good reason to support this contention. Yes, NFB does increase odd order harmonics but the levels are low, and I doubt if anyone can actually hear them reliably. I've found most tube amps with no feedback to have less controlled bass, less dynamics, and subtle softening of the highs.
very interesting article. So mass market audio companies are watching out for their best interest buy supressing this information? If they switched to this better sounding amplification, it could ostensibly eliminate many speaker companies and even their own speakers from being useable? Thanks for finally putting an end to this mystery of why my tube amp was sounding dull. I brought it to a shop yesterday and heard it come to life with appropriate speakers. thanks again and very very impressive website. I look forward to hearing some of those amps one day.
Jmacinnis, The issue here is how the amp will drive the speaker if it has no feedback. This article will explain what is going on:
http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

IMO/IME amplifiers that operate without feedback are more musical as they make less odd ordered harmonic distortion. Since our ears use odd ordered harmonics to figure out how loud a sound is, this is arguably one of the more important human hearing rules! The use of feedback violates this rule.

However the audio industry does not like to talk about this, and is something that separates the high end audio community from everything else; we talk about it occasionally instead of never :)

Ultimately you have to be more careful of your speaker choice; this is doubly important because a speaker that requires the amp to have feedback will never be able to sound like real music- it will at best sound like a good hifi.
I had a Butler TDB-5150 (Tube/MOSFET hybrid multi-channel amp) that had no global negative feedback. It had these specs.:

Freq response: 20Hz to 20kHz (+/- 0.5dB)
Power Bandwidth: -3dB, 50kHz

So I would presume that not having feedback will not impose any audible frequency limitations...

-RW-