I have found amps with less or no feedback sound loose and diffuse with less dynamics... So why this myth perpetuated by audiophiles and even many manufacturers?You did not hear the amp sound loose, you heard the amp paired with a specific pair of speakers sound loose. With a different pair of speakers you may have heard something different.
Dracule1 (Reviews | Threads | Answers | This Thread)
69 responses Add your response
Dracule1, feedback is a very good think if you know how to use it. It improves practically everything - reduces distortions (THD, IMD), lowers output impedance*, widens bandwidth. The only problem is that amplifier presents delay to signal. When you feed this delayed signal back to input stages it doesn't subtract properly - especially when signal is changing fast. It causes overshoot after transition since gain for tiny moment was higher (feedback was late = open). It is called TIM (Transient Intermodulation) This overshoot in time domain is equal to extra odd harmonics in frequency domain. Since odd harmonics carry loudness clues we're very sensitive to them and hear it as bright unpleasant sound. In extreme case when amplifier has very deep global feedback fast transitions might even choke output transistors that stay that way for a moment (charge trapped at the junction) resulting in tiny gaps that are inaudible since brain "fills the gaps" but makes us tired. It sounds unlikely but they designed SS amps like that in 70s before TIM was discovered.
Every amplifier has some form of feedback but it is better if it is local (around one stage) in few places than one global going from the back to front. Also sane designer would make amplifier as fast as possible to reduce delay and would limit bandwidth at the input. I would also be very careful with amount of feedback - tiny overshoot might become 10x higher with 20dB deeper feedback. Reducing harmonic distortions, by the feedback, below 0.1% might not be audible but brings risk of TIM. Also reduction in output impedance might not be necessarily a good thing if your speaker is already over-damped. Class AB amplifiers often have a lot of feedback to linearize output transistors and keep gain constant when both transistors are conducting (gm doubling).
Preamps most likely have stages with local feedbacks. Bandwidth is also easier to obtain with practically no gain and no power stage.
* - Some time ago I made example of feedback reducing output impedance:
Kijanki explained it pretty well as I understand it.
The argument against negative feedback is based on it being an imperfect process that introduces time delays and artificial harmonics accordingly as explained and that this inherent negative can outweigh advantages of applying negative feedback otherwise.
Audiophiles hate anything that is not perfect so being inherently impossible to implement perfectly, negative feedback is a common target of ire.
You might say that the use of global negative feedback causes the amplifier to violate one of the fundamental rules of human hearing/perception: how we determine how loud a sound is. We do that through analyzing the odd ordered harmonics rather than processing fundamental tones.
So if the amplifier has trace amounts of the 5th, 7th and 9th harmonics added, it will not only sound louder than real music of the same volume, but bright/less relaxed as well.
This is one reason why two amps can measure perfectly flat with the same bandwidth on the bench, but one will sound bright and the other won't.
As Kijanki pointed out, global feedback reduces 'output impedance' (I put the term in quotes because it is a definition that is only used that way in audio!). So amps without feedback will have a higher output impedance. If coupled with a speaker that demands a lower output impedance of the amp, tonal aberrations may result.
So you can see that a more ideal combination might be such an amplifier with speakers designed to work with higher output impedances. Then you get proper tonality coupled with no violation of human hearing rules.
For more information see: http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php
I don`t agree with the myth description.It just depends on the amplifier`s chosen output devices and the particular speaker being driven.What good is it to reduce 2nd order harmonics to extreme levels and then replace it with odd high order harmonic distortion(worse),that`s a bad trade off.A tiny amount of this NFB odd order distortion is not natural to human hearing and is processed as irritating and artificial.
Some speakers do require amplifiers with NFB. I`d prefer speakers designed that don`t 'need' these types of amplifiers in the first place.Amps without NFB sound more natural and realistic in my own experience, YMMV.
It is not only 2nd harmonic, but also Intermodulation Distortion, Bandwidth, Output Impedance - practically everything. The key to avoid TIM is to limit bandwidth at the input to one that amp had before feedback (that extended bandwidth) was applied. This will be OK if amplifier had wide bandwidth to start with, otherwise reduced bandwidth will cause phase shift at higher frequencies and wrong summing of harmonics. Also, as I mentioned before, every amplifier has some form of NFB.
"I have found amps with less or no feedback sound loose and diffuse with less dynamics... I know you should design am amp with excellent open loop gain before applying feedback. I can see the use of no negative feedback for low level amplification (eg, preamp and gain stage of CDP or DAC). So why this myth perpetuated by audiophiles and even many manufacturers?"
I don't see why you would say that 0 feedback components better SQ is a myth. If you can, try to audition some Ayre products. That will probably change your mind. (There's other great sounding 0 feedback electronics other than Ayre. I recommend Ayre because I feel that they are the best example.
Thanks everyone for the informative responses. In regards, to speaker/amp pairing, I have three sets of speakers, two pairs are dynamic and one pair electrostatic. In all three of these, overall I liked feedback over none.
As for DHT amp using no feedback, I have yet to hear one that has the dynamics, bass extension and tightness of a KT88/6550 using feedback...that is after taking account power output of the amp. But good DHT no feedback amps have glorious midrange, though not accurate IMO.
Regarding odd order harmonics being irritating and having increased perceived loudness, this is very difficult to pin down in one's audio system among all the other forms of distortions inherent down the chain, unless you are willing to deliberately increase the odd order harmonics in your amp and listen for the change. What I thought was irritating odd order harmonics arising from my amp using feedback was actually distortion arising from the source and boundary effect from room acoustics.
I have two push-pull class AB amplifiers. One is 40 watt/ch with EL34 tubes and the other is 100 watt/60 watt(UL/Triode).The bigger amp is one of the best PP amps I`ve heard it uses 6550/KT 88/KT90 etc. This was my main amp,top choice for driving my Coincident speakers. Three years ago I purchased my 300b SET 8 watt amplifier.From day one fresh out of the shippng carton this SET put the others into storage. It was simply better across the board.
The 100 watt amp has more bass weight and impact(the gap is`nt large however).In terms of tone,harmonics,noise floor,presence,3 dimensionally,nuance and dynamic ebb and flow the SET is superior.When playing high energy big band jazz ,the ability to clearly seperate and hear the individual instruments is extraordinary.The scale,energy and dynamics completely fill my room, it really as if the walls vanish(I`m right there at the venue,spooky real). In the sense of realism,believability and natural sound it`s the best I`ve heard so far.Some SET amps are better than others.The better quality DHT SETs are the most live-like,realistic and trueful amplifiers I`ve ever experienced. This was a profound moment for me as it put me into a different sonic realm,there`s no going back.I`ve own SS (Symphonic Line amplifier) it was`nt close to this.My 100 watt PP was better than it in terms of live flesh and blood presence.The 300B SET is a new standard however.
Just an example of how people have different experiences and impressions. The good news is there`s an amplifier topology out there for everone. SET,PP,OTL,SS,class D, take your pick.What I have discovered with SET I know others have experienced with other types of amplifiers also.
So here's a question.
Can an amplifier with no or minimal negative feedback in play make a CD recording that might be irritating otherwise normally not irritating? Not all recordings are created equal. SOme are more irritating than others for various reasons. I'd say only a very small % of my thousands of CDS, maybe a dozen or less, mainly in the hard rock/heavy metal or a few electronic dance genre MP3 tracks, sound as Atmasphere describes a negative feedback amplifier in practice. I listen to all my CD tracks queued up randomly, so I do not know what I will hear next, yet there is that occasional REM or MEtallica CD that will play and be hard to listen to, but most of the rest seem to sound more like what Atmasphere describes for non NF amps. SO I am not hearing the problem at least, FWIW.
THis is the case even with my Dynaudio Contour monitors, which tend to have a significantly hotter sound running off my gear in their room than say my more laid back OHM Walshes in others.
Can an amplifier with no or minimal negative feedback in play make a CD recording that might be irritating otherwise normally not irritating?
Mapman, you basically asking if it is possible to make amplifier bad other way than NFB. I'm sure it can be done 100 ways from bad power supply to wrong bias current.
NFB is producing TIM only when overused. We're talking global NFB since local ones are almost always there. Transistor amplifier output stage represent voltage source. For that it needs to be regulated = NFB.
In my experience the answer to your question is no. The one quality I`ve come to admire over time with my SET is honesty. What ever is the inherent sound character of the CD that`s what I`ll hear,flaws and all. Some SET amps could perhaps'pretty up' the sound but my does`nt do that. I have some recordings that are thin,brighter and in some cases edgy and that`s how they will sound.But those that are recorded well,oh my goodness.
My SET has less editorializing than all of my previous amplifiers. I reconized this aspect early on, It just seems to preserve the signal with minimal influence.The effect is heard as pure and unfiltered.
I'd say pretty much the same thing with the ClassD Icepower BelCanto ref1000m amps in my system, used with any of my speakers, especially the OHMs and Dynaudios.
My small Triangle Titus XS monitors, which are more efficient and known to be more tube amp friendly, are the ones where I think I could improve the sound with another better matched amp. I do not think use of negative feedback or not alone would be the main factor to determine which amps would work best with those, though I am sure some that do not use extensive NF would sound very good indeed.
I have a pair of Tube Audio Design Hibachi monoblock amps that do a very competent job with the OHMs and Dynaudios (though I prefer the Bel Cantos with those overall), but are probably a better match to the Triangles on paper.
The amp/speaker match is why I dropped this link earlier:
Mapman, if you have speakers that your class D amps are really happy with, there is the chance that those speakers may not be compatible with tubes.
Dracule1, 'tightness' in the bass is not a function of real music but is a function of an over-damped speaker. IOW its an audio system artifact.
Kijanki, I usually do not include degenerative feedback when I make my comments about the negative effects of loop feedback. This is because degenerative feedback occurs simultaneously with the signal i.e. there are no propagation delay issues.
I think it should be pointed out that the use of negative feedback to reduce IM distortion is a bad move. We get very low IM figures without using feedback- IM has a lot to do with power supply design, grounding and parts quality. I am of the opinion that is much better to design the amp to have low IM operating open loop, since IM will occur at the feedback nodes of the amplifier, resulting in a harmonic noise floor (rather than a natural hiss noise floor caused by component noise). The human ear can hear about 20 db into a natural hiss noise floor, but if there is a harmonic noise floor caused by IM (with harmonics and in-harmonc distortions extending to the 81st!), the ear will not hear below that point at all due to the way the masking principle works.
This is one way to get more detail out of an amplifier, since a harmonic noise floor will block the ear's ability to hear detail below that point.
since IM will occur at the feedback nodes of the amplifier, resulting in a harmonic noise floor (rather than a natural hiss noise floor caused by component noise).
Atmasphere, I'm not sure I understand it. IM is caused by the presence of two frequencies on non-linear element resulting in "harmonic noise floor" meaning that there will be new frequencies of small amplitude harmonically related (sum and the difference) to original frequencies. NFB reduces IM by linearizing circuit - therefore REDUCES harmonic noise floor. Very deep NFB can make it almost perfect practically eliminating IM at the cost of introducing TIM. Eliminating NFB might be possible with the inherently linear tubes but not with transistors that require linearization and regulation.
Interaction with the speaker might have more to do with the type of the amplifier - SS representing Voltage Source and Tube amp representing Power Source.
Your explanation of the natural noise floor compared to the 'harmonic noise floor' sounds absolutely right,here`s why. I know my former Symphonic Line and my PP tube amplifiers will measure lower distortion than the 300b SET amp. Yet when I listen to familiar music with these amps the SET in reality had the lowest noise floor for actual listening. It teveals more nuance,inner detail,ambience clues, all the subtle sounds that were either buried or not heard at all with the other amplifiers.This contrast is very apparent. Thanks for this explanation. I could hear all of this easily but did`nt understand why.This is likely a major factor why evrything sounds substantially more real and convincing with the DHT SET in place.It all makes more sense now.
Getting back to my original point...here's a thought experiment. Let's say you design a tube amp with different levels of negative feedback, let's say in increments of 3 dB from 0 to 30 dB. Assuming you have a good speaker/amp match, what you you hear as you increase the negative feedback from 0 to 30 dB? This may be simplistic, but has someone actually performed this kind of experiment?
Dracule1, I have played string bass since 7th grade. 'Tight' is one thing is isn't. Energetic- yes, detail- sure- its a real instrument. But you won't ever find one sounding 'tight'. We may have a semantic problem here; for me 'tight' is punch but little else. Sure, I want the impact, but I want the detail too, and that is something that lots of feedback robs from the bass frequencies- things 'stop' too quickly. IMO/IME its the bass detail and ambiance that goes away first as things go wrong in a stereo.
Kijanki, I would agree with you regarding what NFB is **supposed** to do, but if you also have propagation delay in the amplifier there is no way that the NFB is not mixing with a different frequency- thus the IM. You might want to read Norman Crowhurst- he mentioned this very issue in some depth about 55 years ago.
In a nutshell, low IM is a function of linearity in the various circuits of the amp. If there are non-linearities and NFB is applied, its not reasonable to expect that there will be no IM afterwards. Instead, while the IM will appear to go down, you will find that the energy of the distortion is spread out over the spectrum- that is to say it is by no means eliminated.
Chaos Theory does apply here. If you analyze an amplifier operating with NFB it basically is a chaotic system, complete with bifurcation (which we audiophiles call distortion) and a strange attractor (which interestingly, Norman Crowhurst graphed before Choas Theory was a recognized science!). The formula for NFB and a classic Choatic system are strikingly similar, if not identical.
When you use Choas Theory to analyze an amplifier, then it is easy to see how IM and NFB interact. Imagine a balloon on the floor with air in it, and then a weight placed on top of it; the balloon will squish out to hold the air before it bursts. An amplifier with NFB is similar- when you look at the open loop spectra, using NFB is like adding the weight to the balloon. The spectra expands across many harmonics, with inharmonic information added due to intermodulations at the feedback node (I am nearly quoting Crowhurst verbatim here but that is the succinct way of putting it). IOW the energy of the distortion does not go away nearly so much as we have been led to imagine in many school classes!
The problem with too much focus on negative feedback is that not all amps that use it are created equal. They do things differently and to different degrees.
I have heard some including mine deliver all the goods including bass quite well and naturally for both acoustic and electronic instruments, semantics aside.
Most are probably not able to discern from technical specs a good implementation versus a not so good one. One has to trust their ears. Generalizations focused on one of many design principles that go into these things are of little value IMHO. There is no consensus on this, even among experienced EEs and amp designers, obviously. So try various amps with various designs for oneself and see. Or just go with your gut but be sure to always focus on technical synergies between specific amps and speakers if you do, otherwise all bets are off.
Atmasphere, It depends how tight is tight. Acoustic bass is the tightest type because of the very longest scale. On the other side of the spectrum is Paul McCartney's Hofner Violin bass that has very short scale and horrible definition. I believe McCartney started using longer scale Rickenbacker first time in "Paperback Writer" and it shows - much tighter bass. The best recorded bass (very tight) I have is on Chick Corea "Akoustic Band" and it is upright bass. Perhaps at 7th grade you couldn't afford good instrument?
"the ears have the final say"
This is right and the reason we`re all stating our preferences.I won`t argue with what you hear and say you`re wrong,we just differ based on personal experiences.
As I said in an earlier post,that`s why there`re numerous types of amplifiers in the high end market place.
I found DHT SET no NFB the better choice based on my ears. Your ears led you to PP pentode with NFB,The good news is we`re both very happy,choice is wonderful.
The bass debate may just be semantics regarding the term 'tight'.Two nights ago I had the pleasure to see(and hear) the Bill Charlap trio. Peter Washington was playing the acoustic bass. the club is intimate and unamplified, I was within 10 feet of the stage. The sound of peter`s bass was just beautiful,very full,round, dense and there is much a sense of bloom.I did`nt hear tight or taut,(it was`nt loose and sloppy either, but it was a bit'fat')even when he played very up tempo. The notes seem to linger with much substain and decay, just beautiful live and up close.I do believe there are audiophile qualities/expectations that appear to vary from the reality of live acoustic instruments. If some audiophiles were blind folded and heard peter`s bass playing(but told they`re hearing a system and judge it) they might say it lacked tightness and was too warm and round.People like what they like,but many audio components tend to thin and make the sound leaner(tighter?) than real life presentations i.e. fuller tone and body with weight and presence.
I thought KT88s are beam tetrodes, not pentodes?
I do like the sound of DHT amps without feedback in the right system. They have a certain magic to the sound that I think is missing in tetrode or pentode amps with feedback. IMO, DHTs are more expressive than life making tetrode/pentode comparably "dull", hence I can see why they are popular. However, I don't think DHTs are as true to life...I know many feel just the opposite.
KT88s are beam tetrodes as you state. They are grouped and interchanable with the power pentodes(6500,EL 34,KT 90.120 etc.). Many amplifiers(such as mine) allow use of most of these, especislly KT 88 and the 6550 tubes.DHTs are a completely seperate catagory altogether.
Push pull amps can sound excellent(my Bella Extreme 100 monoblocks built by Bill Baker and my friend`s VAC Phi 300.1 monos). There`s just a different character to the sound.
A 300b SET has 'very little' heat output.There`s only 1 output tube(8 watts) per monoblock.My PP amps run much warmer with 4 6550 per side. You may never prefer a SET no NFB compared to PP with some NFB. I`m just giving my own preference(after living with both amplifier types with extensive direct comparisons) and by no means making any sort of proclamation that applies universally.I wish you continued enjoyable listening with your system.
If I may chime in here on the "tight" bass issue. Atmasphere's description of the string bass is a good one - no orchestral bassist would want to be told that he sounds "tight." I am sorry to say that Kijanki's post in response makes almost no sense from this standpoint. No truly great sounding instrument sounds "tight," though one with a problem, or a bad quality one might. This would be considered a very negative description.
Charles1dad makes a good point: "there are audiophile qualities/expectations that appear to vary from the reality of live acoustic instruments. If some audiophiles were blind folded and heard peter`s bass playing(but told they`re hearing a system and judge it) they might say it lacked tightness and was too warm and round.People like what they like,but many audio components tend to thin and make the sound leaner(tighter?) than real life presentations i.e. fuller tone and body with weight and presence."
To this point, I would add that we also need to distinguish between amplified and unamplified acoustic bass - as soon as amplification is used, as it almost always is in live jazz, for instance, this results in an artificially boosted bass, and a very different sound from the unamplified string bass.
There are a great many audiophiles who do not listen to classical music even on recordings, let alone live, and therefore really don't have any idea what an un-amplified string bass actually sounds like live. Their concept of how bass is supposed to sound is therefore entirely based on either electronically produced or at least amplified acoustic bass. This is the biggest reason why there is so much debate about this in the audiophile community - there are two VERY different references going on. When orchestral musicians use the term "tight," they are never describing timbre. You simply would never hear someone say "He sounds tight!" Instead this term is used to describe how rhythmically together the group is playing - as in tight or loose ensemble.
Thanks for your comments and viewpoint.I made it a point to be clear about the enrivoment in the jazz I attended.Kenny Washington`s stand up bass was 'natural' and unboosted. Definitely not'tight' in the audiophile sense.This is why IMO tube amplifier bass sound more real than most solid state when reproducing acoustic bass, there`s no artificial'slam' added.I do appreciate though that many do like that 'slam' factor and I`m likely in the minority..
Learsfool, Upright bass has better "definition" related to fact that strings are, being long scale, at high tension. Sound of plucked string instrument can be defined by factors like Presence, Projection, Sustain, Separation and Tone. Upright bass has huge projection and great separation making for punchy tight sound, but at the same time has good presence and long sustain. Sound depends entirely on the player, that can play it to use projection and shorten the notes to kill sustain or can play softer because of good presence and use sustain "filling" the room with bass that reverberates. All I'm saying is that short scale bass guitar like McCartney's Hofner will sound flabby no matter how you play it, because it is extremely short scale bass with very low string tension. For the same reason Strat electric guitars sound punchier than shorter scale Les Pauls.
Learsfool, this is all semantics. I have been playing classical guitar since high school, and have dabbled in piano and sax. I have attended concerts at some of the finest halls in the country (Symphony Hall in Boston, etc) and listen to live unamplified acoustic music on a regular basis (Jazz, folk, and classical). By tight, I don't mean bass that has been stripped of harmonic content and sounds dry. I mean muddy bass that has been stripped of harmonic content distorting the timbre of the instrument. Tight bass has initial fast transient attack followed by natural decay and rich harmonics. This can be heard with plucked stand up bass, low piano notes, kick drum, tympani etc. There are exceptions of course such as wind instruments, like church organ, tuba, etc.
As an aside, many years ago I had LWE speakers, which provided negative feedback from the speakers back to the amp. The idea was to monitor the speakers response to the input signal and correct the differences by providing feedback to the amp. So, it was an attempt to correct the imperfections of the speaker, rather than those of the amp. As I understand it, the feedback was trying to do more of a long term, overall correction rather than a short time duration correction. I was never sure how much the feedback effected the sound, but I did like them paired with an old Dynaco SS integrated. People always asked why I have wires coming out of the top of the amp. Obviously, the idea never made the main stream, but the speakers were well reviewed at the time.
I listen to all sorts of music and even play electronic keyboards in my band.
'Tight' to me in its simplest definition is an artificial coloration imparted by overdamped speakers. You get a fast attack, but not so much body behind the initial thump. Depending on the amount of overdamped issues (no speaker is made that needs more than 20:1 BTW) this amount of body is variable. What I find is the low frequency ambient signature of the room is the first fatality to this problem.
Any tube amp can be made to have a 20:1 damping factor or more with enough feedback. Most transistor amps have considerably more. What I am talking about here is not really saying that the amp can't play bass right, but if there is no speaker that is not overdamped with that amp then its a moot point.
The head engineer of EV wrote a 2-part article about this back in the late 50s. You might think that somehow the physics that he was writing about went away in that time, but they didn't... about the only thing that is really different is that there are 4-ohm speakers now. If we are talking about a 4 ohm speaker, then the damping factor of the amp can be up to 40, as damping factor relates to 8 ohms only during measurement, whereas any speaker can be overdamped if its impedance is more than 20X that of the amp.
Ralph, My previous speakers had smaller woofers and smaller cabinets but bass extension was 5Hz lower. Bass quality wise it is much better now with more natural attack and decay. It sounds more even, resonating less with the room (same amplifier). I read that speaker bass can be tuned for max extension (it sells!) or for the lowest distortion. Bass now seems to be very melodic, natural and effortless while before it was a little congested. As for damping, my amp has DF=4000 at low frequencies, but it doesn't make much difference since xover inductor in series with the woofer is usually about 0.1ohm limiting DF to 80.
Dracule1, on p4 of the Stereophile article they test Cary amp with adjustable feedback and describe sound change. At the end no feedback sounded the best. It doesn't mean that amp had zero feedback. There could be still a lot of local and perhaps even some global feedback left.
Hi Kijanki - yes I realize this is semantics; I do now understand what you were talking about - however, your use of terms is very bizarre to an orchestral musician. Orchestral string players do not speak of the length of their strings as "scales" (and neither do they use the term "tight" to describe their sound, as I explained in my earlier post). The use of the term "scale" in the way you do must be a guitar thing. Probably because the length of the string on an orchestral instrument is always the same, whereas in a guitar it would not necessarily be.
I also fully understand the way in which some audiophiles use the term "tight" - I just disagree that it is a good term. I could see the term applied to an amplified string bass or acoustic guitar sound, but this would be because of the amplification, not because of the instrument itself. The term is simply not used in describing live, un-amplified, acoustically produced music (certainly not in the classical world, anyway), and therefore doesn't really have anything to do with Harry Pearson's "absolute sound" concept, if one agrees that that should be the standard for what a system should ideally sound like. If one doesn't agree, then by all means use the term. Dracule's definition of the common audiophile way of using it is as good as any, though I have never heard a pianist use it. Drummers, yes. Even timpanists, though they generally use the term to describe the tightness of their drumheads and the effect this has on the sound, which is somewhat different, I think, from Dracule's description.
Charles1dad, I agree with your post in response to mine completely.
Learsfool, Term "Scale" applies to all string instruments. Please read here: http://en.wikipedia.org/wiki/Scale_%28string_instruments%29
Upright basses have 10" longer scale than bass guitar. It makes for much higher string tension* resulting in (as Dracule1 called it) "initial fast transient attack followed by natural decay and rich harmonics".
*Since string tension goes in square of length 43.3" upright bass has twice the string tension of McCartney's 30.25" Hofner bass that in comparison sounds flabby with lack of definition - poor attack, muddy overtones. .
Learsfool, I know "tight" is not a term most musicians use. But we audiophiles (and closet musician on the side such as myself) have our own set of terminology that will befuddle most musicians. We use many different terms to describe the similar, if not the same, thing (eg, instrument tone, tonality, timbre, overtones, harmonics).