|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
77.43.73.35
In Reply to: RE: Have you seen this? posted by Ralph on October 05, 2010 at 13:03:52
If I understand correctly, Ralph points out that the most desirable speaker/amplifier combination is the one that gets closer to the so called 'power paradigm' operation, that is, when the amplifier/speaker combination delivers the same ACOUSTICAL power irrespective of the speaker ELECTRICAL impedance.
As the acoustical power irradiated in the ambient is always a very tiny fraction of the electrical power delivered by the amplifier to the speaker, this makes sense to me and I could not agree more with this vision.
Then Ralph states that, when approaching 'power paradigm' operation, negative feedback usually creates objectionable distortions and that 'zero feedback' is better.
Also this makes sense to me, as any feedback, by its inherent automatic control of the amplifier output impedance, certainly interferes with 'power paradigm' operation, that, on the contrary, would require precise matching of amplifier output impedance and speaker input impedance (in order to get a flat acoustical response).
Unfortunately it must be taken into due account that, to my best knowledge, 99.999999 % of the available commercial speakers are designed in such a way to guarantee a more or less flat ACOUSTICAL frequency response (in anechoic chamber) when driven by a constant voltage source.
As one of the practical ways to force an electrical amplifier to behave like a more or less constant voltage source is by using some negative feedback, this means that 99.999999 % of audiophiles, because they buy (or they own) standard commercial speakers, they are forced by the 'audio status quo' to be quite reluctant to completely abandon feedback.
Simply because if you own standard commercial speakers, you are likely to trading in 'zero feedback' in exchange for 'tonal imbalance', as the ambient acoustical frequency response is going to be less flat than with some carefully selected feedback.
In conclusion, I admire that 0.0000001 % of audiophiles that can afford the luxury of purchasing the right amplifier/speaker combination (with zero feedback, of course) able to deliver a flat acoustical ambient response (according to 'power paradigm' operation).
Those fortunate people will certainly experience the 'less objectionable distortions' on earth but please, don't recommend to the vast majority of other audiophiles to 'blindly' resort to the supposed 'zero feedback' miracles.
This in my opinion can not be considered correct information.
Best Regards
Luca
ecc230
Follow Ups:
The idea is that if an amplifier has no feedback and a high output impedance, it becomes a power source rather than a voltage source.
SETs are an example of this, as are Nelson Pass's First Watt amplifiers.
Speakers that are designed around the Power Paradigm include all ESLs, nearly all horns (except the Avantgarde Trio), 'full range' single-driver speakers, and a good number of bass reflex and acoustic suspension speakers (Coincident Technology is a good example).
Note that the impedance curves are not flat- that is not a requirement of the Power Paradigm. What is happening here is that the Power Paradigm is saying that distortion products are a greater source of coloration than frequency response is. However it is possible to take advantage of the amplifier characteristic in speaker design, some of Duke LeJeurne's (Audiokinesis) designs are examples of that.
IOW, this is a fairly large portion of the high end market, where that not the case, there would not be the equipment matching conversation that is everywhere in high end audio.
"distortion products are a greater source of coloration than frequency response is"
The influence of frequency response depends on listening level.
Because of ear sensitivity, for low level listening a 'narrower/less flat' frequency response is acceptable, but this is not true for faithful orchestra reproduction.
For the interested readers I recommend to take a look at Chapter 14 "Fidelity and Distortion" of the Radiotron Designer's Handbook (IV edition, downloadable at the Peter Millet site).
In my opinion your point of view would be barely acceptable only in case of low level listening, because distortion products are usually vanishingly small and hence frequency responce can be somewhat degraded.
To summarize our discussion:
According to the proposed 'power paradigm/zero feedback' approach,
- on one hand we should accept a less than optimal (and variable with different speaker types) frequency response because distortion products are a supposed 'lower source of coloration',
- on the other hand, due to the extremely low power transfer to the speaker, we will certainly get the highest distortion products (because we are forced to push harder the output tubes for getting and adequate acoustic dynamic range).
Power paradigm now looks to me what it should look like to all un-biased audiophiles: a minor niche of high-end market of small or no interest for the average people.
Best Regards
Luca
ecc230
" on one hand we should accept a less than optimal (and variable with different speaker types) frequency response because distortion products are a supposed 'lower source of coloration',"
In todays audio landscape it is a relatively straightforward operation to fix frequency response. A decent digital equalizer will take care of this nicely (The Behringer DEQ2496 is a good choice that can be inserted between transport and DAC)
It has also been long apparent to me that listeners are far more tolerant of linear distortions (i.e. frequency response) than they are of non-linear distortions (cone flex and breakup, cabinet resonances etc.) and the same holds true for electronics. People mistakenly make a big deal out of 1 db FR variations (potentially depending on the speaker impedance curve) from a SET amp not but the bigger damage to real natural sound, NON-linear distortions and in particular high order non-linear distortions. This is what negative feedback brings to the party and its effect on the sound can be readily heard (even 2db worth according to the initial poster).
I have been experiencing the best sound of my life once I removed all negative feedback from my system. My amps don't have it, my DAC doesn't have it and my phono stage doesn't have it.
In fact, the total number of active stages has been minimized as well. There is only one active stage in the DAC (the tube output is taken directly from the DAC chip output) and three active stages the integrated amp (one JFET, one MOSFET and one big output tube). The whole system runs in Class A. The phono stage adds 1 more active gain stage but still no feedback. As a result it is about as harmonically complete and dynamically lively as I have heard from any system. I use digital equalization in limited amounts and this gives a FR accurate room response.
A big "amen" to just about everything you said up there.
I know about Fletcher-Monsun, this is not what I was talking about. I **was** talking about how the ear perceives distortion as a coloration, for example distortions of only 1/100th of a percent of the 5th, 7th and 9 th harmonics are perceived as brightness, despite the frequency response of an amp with such distortion as being otherwise flat.
The same is true of low-ordered harmonics will make the amp sound 'warm' or 'full', perhaps even bloated in the bass, despite flat frequency response.
Some CD players can exhibit so much inharmonic distortions that no amount of turning down the treble control can tame their brightness.
Do you understand now my point??
But I prefer to stick to audio science.
My impression is that you seem to escape any meaningful technical discussion by never making clear your points.
You look too often so vague, so far away from common technical vocabulary, always trying to referring to something else (that is not the topic of the matter under scrutiny).
I won't follow you on this terrain anymore.
If you want to positively contribute to this forum discussion please stick to the points at hand, don't ever assert: "low-ordered harmonics will make the amp sound 'warm' or 'full', perhaps even bloated in the bass, despite flat frequency response".
This is basically nonsense as long as you do not put numbers there.
And again, what the hell your statement "some CD players can exhibit so much inharmonic distortions" has to do with 'power paradigm high impedance"?
Which CD models?
Who is the manufacturer to be blamed?
But who was ever asking your opinion on CD players?
Who cares?
In my experience CD players are excellent audio sources.
My point is clear: masked advertisement maybe good for you but it is extremely disturbing for me.
Best Regards
Luca
ecc230
I am assuming that it has something to do with language.
Had you been following along, then I doubt you would have said this:
And again, what the hell your statement "some CD players can exhibit so much inharmonic distortions" has to do with 'power paradigm high impedance"?
So I am assuming that either you did not read something, or read it and misinterpreted it based on a misunderstanding of the language.
To flesh things out a little, General Electric did a study in the 1960s, wherein they found that humans find odd-ordered harmonics to be objectionable, and very small amounts, less than 0.1% . By contrast, the same study showed that humans may not even notice lower ordered harmonics (2nd,3rd, 4th) until they get over 30%! The former is easy to demonstrate with a sine/square generator, an amp, a speaker and a VU meter.
If you want to positively contribute to this forum discussion please stick to the points at hand, don't ever assert: "low-ordered harmonics will make the amp sound 'warm' or 'full', perhaps even bloated in the bass, despite flat frequency response".
I *thought* this was common knowledge, and germane to the conversation- I refer you to Norman Crowhurst.
What a pity. It's not a language problem, it's you that are likely running out of meaningful argumets for defending the supposed advantages of 'power paradigm/zero feedback'.
As in your recent 'clash' with Tre' you escape again and again the technical points and start referring to something else.
As far as the argument "General Electric did a study in the 1960s, wherein they found that humans find odd-ordered harmonics to be objectionable, and very small amounts, less than 0.1%" is concerned, please re-publish some relevant excerpt of this 'unknown/unpublished' study on this (or your site), so that we can all check what the study assumptions were at the time (istead of relying on your statements only).
And as odd ordered harmonics seems so dangerous according to this GE study I will be glad to see the measured distortions of your audio products, let's say at 12 volt peak across ESL loading, 330 Hz, just to see if the odd-ordered harmonics fall below 0.1% as you consider so much essential.
Best Regards
Luca
ecc230
Luca, Tre and I figured it out even if you didn't.
Now, since you don't believe me, you now have to prove me wrong. This is an easy test. Get a sine/square generator, an amp (any functioning amp will do) a speaker and a VU meter.
Set the generator for sine, at 0 VU and listen to the result through the speaker. Now cover up the meter. Turn the volume down, set for square wave and bring up the volume until it sounds as loud as before. Now observe the meter. You will see that it is a good ways down from the sine wave- probably about 20-25db.
This is because we use odd ordered harmonics to see how loud a sound is. A square wave is all odd orders. So naturally it sounds a lot louder than a sine wave.
The fact of the matter is I have not been changing the subject- instead I get the sense that you are actively trying to not understand. But you need to perform the test above!
I see a serious efficiency problem and an 'out of control' ambient frequency response.
I a recent post of mine
http://www.audioasylum.com/forums/otl/messages/3/33624.html
I recalled the classical graph of (electrical) power transfer/efficiency vs. the relative amplitude of source and load impedance.
Assuming that source impedance is high (w.r.t. speaker load impedance) this means that the amplifier/speaker combination, when operating according to power paradigm, lies at the very left of the peak of the curve of electrical power transfer, close to the XY origin, where efficiency is extremely low.
If this is correct, power paradigm is basically negating the very purpose of audio power amplification.
What is even worst (in my opinion) is that the power paradigm portion of the power transfer characteristic is so steep to actually preventing any tonal balance with the vast majority of commercial speakers (that are designed for optimal performance when driven by a voltage source and whose electrical impedance varies quite a lot over the operating band).
Pushing to the limit your argument, if the source impedance were infinite we would even obtain a flat frequency response (irrespective of any variation of speaker load impedance), but with NO POWER TRANSFER at all to the speaker.
If this is the price of power paradigm I would rather stay with moderate feedback (and voltage source paradigm)!
I still do not know how to faithfully reproduce the orchestra dynamic range (more than 90 dB) without enough 'raw' power.
Best Regards
Luca
ecc230
to some sort of weird extreme. "high output impedance" refers to an amplifier that is perhaps more than 0.5ohms output impedance- some appreciable fraction of the load impedance...
IOW power production is the issue; look at all zero feedback SETs made, they are all part of this.
0.5 Ohm difference is nothing.
English/American is not my mother tongue but I understand very well the meaning of the words.
High source impedance in technical world means many, many times the load impedance.
If the difference of input and output impedance is so small (0.5 Ohm or the like) I see that 'power paradigm' is nothing different from common impedance matching (for almost maximum power transfer).
Nothing new, nothing particularly interesting, in a single word, 'moonlight'.
Good for advertisement, not for engineers.
Best Regards
Luca
ecc230
When you are referring to a conversation about the output impedance of amplifiers, in the US anyway, the engineering language is such that 4 ohms would be considered a 'high output impedance' and 0.5 ohms would be somewhere between low and high, as most amps that have a 'low output impedance' are often 0.1 ohms or less.
There is no such thing as a power amplifier that has an output impedance that is say, 100 ohms or more. That's not an amp that can make any appreciable power, so by definition its not in the conversation.
So it does appear that the mother language is playing a role here.
Ralph...what is it about the construction of the Trio that excludes it from the concept of the "power paradigm"?
The speaker, depending on the version, is built without chokes in the crossover, only caps, and none of them are shunts.
So what happens is in the low frequencies the speaker is about 19 ohms. Then the mid horn rolls in and since the bass horn is not rolled out, the two are in parallel, so the impedance drops to about 8 ohms. When the tweeter rolls in, neither the bass or mid is rolled out, so they are in parallel with the tweeter.
This is why so many owners of this speaker have really had to dig to find a tube amp that will play the speaker properly. The reason the speaker is built this way is that the designer used a transistor amplifier.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: