Hi!
I believe it depends more on the kind of SET amp and how tough is the load (loudspeaker effieciency, impedance and phase angle).
Most DHT SET triode amps have low distortion levels and in general sound quite alright without NFB, when matched to SET-friendly speakers.
However, SET amps based on output pentodes or tetrodes usually require a modicum of feedback to reduce distortion.
If you want to hear the effect of NFB, Cary's 805 SET amp has variable feedback. I have learned that with speakers that present a difficult load, 6 db of NFB is a good thing.
In summary, as with most things in life, there are no definitive answers set in stone, one must experiment and find out what is the best compromise.
Enjoy the music |
Almarro 318B has 6db of feedback.
The best bass I have ever heard from a SET.As reviewers have said it sounds far more like a 100 watt SS amp in terms of bass control.
A superb sounding amp in all other areas too. |
The Art Audio Quartets (845 PP) have a switch for 0 or 6 db of feedback. I was trying them on my Maggies and preferred the feedback initially. One day I had some "girl with guitar" music on and went back to the 0 setting, which was preferred. It was not a huge difference but nice to have the option. Bass was better with, midrange better without.
It's what you said in your last sentence, "just about the sound". Don't get caught up in marketing hype. |
There are many ways to make great sounding amps, and between theory and listening there can be a great gap. Like Onemug says, don't get caught up in what should or shouldn't sound better based on topology or techincal specs, just listen. |
Using or not using feedback is NOT just about marketing hype.It fundamentally changes the damping factor of an amplifier.
Most speakers will not work well with zero negative feedback ampifiers.This is especially true for speakers with low impedences-below 8 ohms[most modern speakers].This is an electrical fact.Adding some feedback increases damping factor and this in conjunction with using tubes able to deliver plenty of current[not 300Bs] will seriously improve bass control.
Of course some people like the looose and bloomy bass of zero negative feed back SETs-but it is not accurate. |
Jtgofish, that is not entirely true, although there is a class of loudspeakers for which it probably is. see: http://www.atma-sphere.com/papers/paradigm_paper2.htmlfor more information. Negative feedback violates one of the more important rules of human hearing by enhancing the odd-ordered harmonics used by the human ear as loudness cues. Hundredths of a percent is quite audible- bright, hard, harsh, clinical, chalky... -these are all words to describe extremely slight amounts of odd-ordered harmonic distortion. |
Atmashere,
I am sure what you say is theoretically correct.
I have had the opportunity to use amplifiers with variable feedback.Experimenting with the amount of feedback is very interesting.Certainly too much detracts from the sound.Most noticeable is the collapse in depth and precision of soundstage when too much is applied.
However to my ears a small amount of feedback gave a much more controlled and natural sound than no feedback.
P.R.A.T. was much better. |
Jtgofish, if you have a speaker designed using Voltage Paradigm rules, then what you experienced is absolutely right. We have had similar experiences.
A speaker using Power Paradigm rules will often work better, including the bass, using an amplifier without any feedback at all. Obviously such a speaker is designed expecting a different set of rules from the amplifier. |
Atma - how do I know if my speakers are designed with voltage or power paradigm? (EgglestonWorks Fontaines) |
Rockadanny, it all has to do with the intention of the designer. In general, if the designer uses tubes its probably a Power Paradigm speaker he created. Voltage Paradigm designers usually use transistors. That will answer the question about 95% of the time. Beyond that, its a good idea to talk to the designer and/or read any white papers he may have written, or simply ask what sort of amplifier he recommends.
The equipment matching conversation that is all over high end audio is a direct result of the conflict of these two paradigms. |
Ralph;is there a speaker that performs well in your opinion that operates within both of paradigms you mentioned;just wanted to hear your view on those types of speakers. |
I prefer less feedback. I've owned amps that had adjustable feedback and they always sounded more musical with zero feedback.
This assumes that you have a reasonable power match between your amp and your speakers. If you don't, then a little feedback may sound like it provides greater control, but it will come at the expense of transparency, harmonics and tone. I prefer to correct the amp and speaker mismatch before I compensate by adding feedback.
In systems where big punch and power are desired more than sublety, I've heard push pull amps that I liked using up to about 6db of feedback. Any more than about 6db of feedback in any amp dampens harmonics and transparency too much for me. An example of an amp using about that much feedback that is very musical is the Tom Evans Linear A. Still, when compared to an excellent zero feedback amp, one gives up some harmonics, decay and transparency.
In short, I don't think an SET amp with more than 5db feedback is a "good thing" and I believe less is better. |
Rleff, the key to that is a flat impedance curve. The Avalon Eidelon does that well, as well as most magnetic planars, whose impedance curve has nothing to do with box resonance :)
There is another issue- the difference in performance between low impedance speakers and high impedance. Actually, the difference in impedance has little to do with the speakers and everything to do with the amps: **all amplifiers** regardless of the technology, will sound better driving higher impedances. IOW if sound quality is your goal, it will be better served by an 8 or 16 ohm speaker as opposed to 4 ohms, all other things being equal.
Another way to put this is if you had a speaker with 2 woofers, 2 midrange units and 2 tweeters, and all the drivers were 8 ohms, it would sound better wired as a 16 ohm load than it would wired as a 4 ohm load, regardless of the type of amplifier involved.
OTOH if sound **pressure** is your goal, and you have a transistor amplifier, a 4-ohm speaker might be preferred. |
Where I termed it "big punch and power" I think Ralph is calling it "sound **pressure**" and I believe his term is more precise. If that is the goal, I'm skeptical that a marginal addition or subtraction of negative feedback will have significant sonic benefit or detriment.
I do believe that when using a properly matched speaker and tube amp combination, that marginal reductions in feedback yield significant sonic improvements. If your goal is to hear the harmonic tones of a violin, be able to pick out the timpani skin from within Solti's 9th symphony, hear how Neil Young's rythm guitarist scrapes the side of the pick along the length of a string or hear the unique tone of Hendrix's marshall amp at any volume, then less feedback is better. Jeff |
Ralph - Thank you. I greatly appreciate all of your contributions to this forum. The equipment matching conversation that is all over high end audio is a direct result of the conflict of these two paradigms. If there was one fact I wish I had known when I started my audio journey it would be this one. |
Gentlemen,
Thanks for all of the great responses. Everyone basically confirmed my own thoughts about NFB. I have a Cary sli-80 all triode w/ zero NFB. Its so much better than the regular sli-80 and they only have about 5db nfb. There is just a huge dif in musicality with mine vs the regular sli-80's. I have been looking at getting a SET amp and I found out the other day that one of them that I was considering has 15db of nfb!!
I think it's good advice to stay with zero nfb amps. I have coincident speakers, so they work great with tubes.
Ralph I want to come to RMAF and hear one of your amps, looking forward to it.
Thanks
Mike |
