Are all amps being built wrong?

Thanks @bifwynne , one problem is that I am 1000-2000 miles from most shops. But I do an appointment with one in 3-1/2 weeks.


My ARC gear has also worked fine… for many decades.

@holmz ... I agree with your last 3 sentences.  As to your 2 bullet points, ... sure, try to spend time educating your ears, switch gear in and out, burn sales people's time.  If the sales people think your not a buyer, they will toss you. 

IME, it takes time to listen to gear, switch stuff in and out, and so forth.  And even then, unless you are switching gear in and out of your system, ... in your home, there is risk that what you think sounds good in the store may not sound so good when you get it home.      

That is why I read reviews and buy brands that I trust, like ARC.  So far, I have been lucky.  But I will never know if I could have done better because I don't sit in brick and mortar stores comparing gear.  I simply don't feel comfortable. 

The business part is that, IMO, it is unfair and unrealistic for one to think that they can sit for hours on end in a brick and mortar store, take up a salesperson's time comparing amps, ... and then not to buy something.

• The sales people are there for that purpose.
  
• And the profit margin is designed to account for that.
  

The alternative is to buy gear and sell it, doing the demos in one’s own home. Or just trusting the internet to provide information on what one is looking for. And that later part is bit problematic with confirmation bias.

@atmasphere  ... thanks Ralph.  I would be fibbing if I said I understood everything you just posted, but at a high level I can appreciate a little bit better why feedback is used and what its limitations are.

Too bad our hobby is really a business.  What I mean by that contradiction is that it would be interesting to compare various types of amps (e.g., zero feedback, some feedback, solid state, tubes) in a controlled environment with the same speakers at the same time.  That is the hobby part of our musical pastime.

The business part is that, IMO, it is unfair and unrealistic for one to think that they can sit for hours on end in a brick and mortar store, take up a salesperson's time comparing amps, ... and then not to buy something.

Thanks again for the great posting on feedback.  It shows that a lot of thought and experience goes into designing and manufacturing really good audio equipment. 

Ralph, thanks for the info. I recall that many years ago you explained that global NF creates temporal intermodular (TIM) distortion because of the infinitesimally small amount of time that it takes for the signal (after phase inversion) to loop back from the output stage to the input stage. That small time delay causes TIM distortion, ... if I recall your explanation correctly.

@bifwynne Not TIM, but HD. You can look at it in terms of propagation delay (easily known and seen in any class D amp) or you can look at it in terms of phase shift. Either way it means that as frequency goes up, the feedback is increasingly erroneous. Eventually you arrive at a point where feedback causes oscillation since it is no longer negative. Put another way, you have two qualities in any amplifier: its gain bandwidth product which describes how much gain is available at a given frequency to support your feedback, and the phase margin of the amp, which is the frequency above which the amp will oscillate if feedback exists above that frequency. (This is why the Futterman amp could go into oscillation if presented with certain loads, since that load would affect the feedback and allow the amp to exceed its phase margin.)


So one conclusion you can draw from this is that at low frequencies you can have a lot of feedback. Its relatively easy to have 60dB at 10Hz; the real question is how much do you have at 10KHz or 20KHz?? Because most amps simply lack the gain bandwidth product, their feedback value falls off with frequency (meaning that distortion also increases with frequency, resulting in brightness and harshness). To support 60dB at 10KHz you're going to need open loop bandwidth of nearly 100MHz (and yes, that's an 'M')... right away you can see that no amp made can support a claim like that. This is why THD is usually measured at only 100Hz, where things are 'safe'. But that practice sweeps the dirt under the carpet because its between 3KHz and 7KHz where the ear is most sensitive (Fletcher Munson) and at these frequencies the amp will simply have less feedback.


Something also going on is that due to non-linearities existing at the feedback node there will be bifurcation of the audio signal created by the feedback. This results in a noise floor composed of harmonic (which are much higher ordered) and inharmonic (intermodulation) noise rather than just hiss as seen in a zero feedback circuit. Norman Crowhurst wrote about this in the 1950s; this is pretty well-known. IME this kind of noise floor is harder for the ear to penetrate while the ear can typically hear about 10dB or so into a 'hiss' noise floor. I suspect this has a lot to do with how well the amp can portray low level detail.


At any rate, if you really want the amp to sound musical, you need two things; the first is that you'll want to see is that the distortion is the same at all frequencies in the audio band. With traditional solid state this is very hard to achieve using feedback. Its also hard to do with tubes. One way around this is to build a wide bandwidth circuit that has good linearity and no feedback. That's been our solution for the last 46 years (as it is for Ayre and a few others). Of course you pay a price for this (as you do for anything); in our case it means that matching the amp to the speaker requires more attention.


Another solution is class D since you can set up a class D amp with so much feedback that it oscillates (because its phase margin is exceeded) and then use that oscillation as the switching frequency. Now you can have enough feedback that even though the bandwidth of the amp may not be that wide, it can have consistent feedback (and thus low distortion) at *all* frequencies with enough of it that phase shift can be as well controlled as a wide bandwidth amp of no feedback at all. This much feedback can also be enough to clean up the mess that feedback normally creates when there isn't enough of it.


The other thing you need to have is a proper distortion signature (something most people call the amp's 'sonic signature'; quite literally this is the difference we hear in all amps) that prevents the amp from sounding harsh due to unmasked higher ordered harmonic content.

@atmasphere 

Ralph, thanks for the info.  I recall that many years ago you explained that global NF creates temporal intermodular (TIM) distortion because of the infinitesimally small amount of time that it takes for the signal (after phase inversion) to loop back from the output stage to the input stage.  That small time delay causes TIM distortion, ... if I recall your explanation correctly. 

Perhaps in the case of "[u]ltralinear" and 'partially cathode-coupled' topology," the physical distance is shorter because the local feedback loop is shorter.  Just a guess.        

Does that type of feedback create the same distortion as the type of NF used in solid state amps or is the distortion still there but tamed to some extent?    

@bifwynne 

It does not seem to but I've not done a lot of research on the topic. But its well-known that ultralinear (if set up right) gives the same linearity as a good triode; it can be treated in terms of linearity as if a triode is in the circuit. The cross-coupled cathode feedback has a similar property of being more of a local feedback rather than loop feedback. So IMO its not harmful.

@atmasphere 

Ralph, quick question on negative feedback (NF) distortion.  I seem to recall that Audio Research output stage coupling is a combination of "Ultralinear" and "partially cathode-coupled" topology.   I have no idea what that means.  I clipped it from a Google search. 

Does that type of feedback create the same distortion as the type  of NF used in solid state amps or is the distortion still there but tamed to some extent?      

Speaker drivers are designed to work best when fed with a voltage source.

I wish it were this simple. The problem has been that the technology hasn't been up to the theory.


To get a tube amplifier to behave as a voltage source you need feedback. But if you don't have enough feedback, one consequence is that higher ordered harmonic distortion will be generated. As seen earlier in this thread, there are very few tube amps that actually had enough feedback.


Higher ordered harmonic distortion is audible as harshness and brightness. You can have a THD of 0.005% and it will be audible. The reason is the ear is more sensitive to higher ordered harmonics than anything else because it uses them to sense sound pressure.


Because most solid state amps need feedback to be linear, they too have this problem (and this is why solid state amps have a reputation for brightness). The Voltage Paradigm relies, for the most part, on amps having feedback (or otherwise a very low output impedance).

Because the ear converts all distortion into some sort of tonality (for example, the 'warmth' of tubes is caused by the 2nd and 3rd harmonic) it also gives that tonality extra attention- so much so that it will favor it over actual frequency response. Add to that the simple fact that no loudspeaker is actually flat and it becomes possible to have a loudspeaker that isn't designed to be driven by a voltage source.

Such speakers were common in the 1950s and 1960s. To identify them, look on the back for a level control or switch that allows you to adjust the midrange and/or tweeter. There are such loudspeakers made today as well. Look on the back of a Sound Lab and you will see that its quite adjustable, to allow it to be compatible with amps that have a variety of output impedances.

Speaker drivers are designed to work best when fed with a voltage source.

**Most** are but some are not. ESLs for example...

If one was to use a DSP, then (I think that) all gets normalised to be right whether using a voltage or current source?


Room correction in modern systems would be doing the same thing I think.

Single malt whisky please.The Scots really dislike it being called Scotch.

I also did not measure their spectral output, just listened. using my favorite measurement aid of the time, single malt scotch.
vidmate  mobdro

Ralph (Atmasphere) posted: 

If you add a resistor to the output of a solid state amp it will indeed simulate some sort of tube amp that uses feedback. Most transformer coupled tube amps with 15dB of feedback will act as a pretty good voltage source. You might have to play with the taps on the output transformer. Between Voltage source and Current source there is Power Source, which is how a tube amp will behave if it has no feedback or if it has voltage and current feedback of equal amounts.

I think Ralph is right on.  My ARC Ref 80S is a transformer coupled tube amp that uses 14db of negative feedback.  And I do **play** with the taps to get the best sound.  At this point, I have settled on the 8 ohm taps, which I believe have an output impedance of approximately 1 ohm.  The DF is 8 if the amp is coupled to an 8 ohm speaker off the 8 ohm taps.

As to comments made by others, I believe that it is correct to say that matching my amp (or one like it) to a speaker that was designed and voiced to be driven by a constant voltage source amp is a bit of a tone control.  To some extent the amp's voltage output will vary to some degree as a function of speaker impedance, which in turn, will vary with frequency.  

All that said, I like the sound of my amp/speaker combo, ... tone control and all.  Ironically, I surmise that between room affects which probably boost the bass a bit and and the tone control affects of the amp which soften the tweeter/high frequencies, the overall frequency response and sound is musical and pleasant.  Just my opinion. 

BIF

All amps should come with variable feedback.

@atmasphere 

Running zero feedback is a nice way to insure stability. What is the output impedance of your amp?

I was talking to a fellow yesterday that said that the feedback of the Prima Luna was too high.

Do you have thoughts on that?

Of course it means bias stability demands attention, but i solved that decades ago.

:)

oops, very late to reply to @atmasphere

Output Z of the amps, depending on model is around .1 - .2 (rarely this high) ohms. Small but very clearly not zero.
And yes the stability is a very big plus.
Of course it means bias stability demands attention, but i solved that decades ago.

Tony - I can fund transcription and or digital copy of the RM/Futterman tapes. i believe those would be important historical context. Advise please.
best
Jim

United States Patent Office. 2,773,136. Patented Dec. 4, 1956. 2,773,136. AMPLIFIER. Julius Futterman, New York, N. Y.

Thanks!

One of my business partners at the time had West Sound Lab monsters and used, of course, my amps. Sounded terrific (all speakers are, of course, to someone’s taste). no issues whatsoever. Just sayin’
Edit, after some coffee, do remember back (oh, 300 posts ago, i get it :-)) that my designs are not zero output impedance. Since they do not use global feedback, there is always *some* meaningful output impedance, but also, pretty much zero chance of instability. Both often issues with large 'stats.

I forgot that the Sound Lab has adjustments on the back to allow for more bass and less treble. Anytime you see that sort of thing on the back of the speaker, its an indication that the speaker does not conform to the voltage rules. It has the adjustments to allow the unknown output impedance of the amplifier to work with the speaker (IOW they are not there to adjust the speaker to the room). So that was not the best example on my part.

Running zero feedback is a nice way to insure stability. What is the output impedance of your amp?

This isn’t quite correct, just so you know. Try putting your amp on a Sound Lab ESL sometime.

One of my business partners at the time had West Sound Lab monsters and used, of course, my amps. Sounded terrific (all speakers are, of course, to someone’s taste). no issues whatsoever. Just sayin’

Edit, after some coffee, do remember back (oh, 300 posts ago, i get it :-)) that my designs are not zero output impedance. Since they do not use global feedback, there is always *some* meaningful output impedance, but also, pretty much zero chance of instability.  Both often issues with large 'stats.

I also did not measure their spectral output, just listened.  using my favorite measurement aid of the time, single malt scotch.

Now would someone prefer quickie's or yours? maybe.

United States Patent Office. 2,773,136. Patented Dec. 4, 1956. 2,773,136. AMPLIFIER. Julius Futterman, New York, N. Y.

This invention relates to amplifiers, and more particularly to audio-frequency power amplifiers employing large amounts of negative feedback to reduce distortion, and is therefore useful in the reproduction of speech and music...
One object of my invention is to provide a low cost power amplifier that does not use an output transformer, and is capable of Supplying large amounts of undistorted power directly to a low impedance load of the order of 16 ohms, such as the voice coil of a conventional loud speaker. Another object of my invention is to provide a power amplifier useful over the audio range of 20 cycles to 20 kc. and capable of utilizing large amounts of negative feedback, of the order of 60 db, without instability.
United States Patent Office. 2,773,136.

@clio09 I'd love to see it!

@jasonbourne52 - It's been a couple years so I would have to look at that Harvard H3 again. I seem to recall 1" long beads, but they could have been rods. They were all around the power tube sockets.

@bdp24 - I'll see if I can dig something up to verify. I keep forgetting those AC circles are still accessible although AC banned Roger.

@atmasphere - I have about 3 hours (cassette tape) of Roger interviewing Julius Futterman (also interviews with Saul Marantz, Sid Smith, and Brian DePalma) that he recorded when visiting Julius in NY years ago. Upon opening the door to his shop Julius asked Roger if he could come back in an hour as he was about to get started eating his lunch (which he brought to work everyday in a brown paper sack). BTW - if you are interested in dissecting an H3 I have two of them. I'll send one off to you if you like. Perhaps we can find out if he was able to pull off that 60 dB of feedback.

I seem to remember Modejeski discussing that Futterman 60dB feedback figure in one of the threads on the Music Reference AudioCircle Forum (though dormant, all posts are still viewable), and maybe in one of his three seminars at The Burning Amp Festival, which I encourage everyone to watch on YouTube. All contain a wealth of free information and wisdom, as does the MR AC Forum.

Julius used hollow ferrite rods on the wiring of his amps - at least the one's I've seen. 

In an effort to solve the oscillation issue in his first prototype and which you noted, one thing Roger found is Futterman himself omitted some things from the H3 schematic, specifically the lack of notation for the ferrite beads

@clio09 

IMO this was intentional. I have a lot of anecdotal evidence from customers that say that the Futterman amps held together while amps made by others using the Futterman circuit didn't. However Harry Pearson recounted an incident where Futterman brought one of his amps to Sea Cliff for audition but before it could be entirely set up, one of the amps went into oscillation and failed (this was in response to a letter to the editor from Harvey Rosenburg in the late 1990s). But all you have to do is leave out one bead by accident and you're sunk. Futterman made most of his amps himself to my understanding.


60dB of feedback in a tube amplifier is an impressive feat! Normally you have such prodigious issues with not exceeding the phase margin of the amp (OTL or not) that most would not attempt such a thing. If 60dB is correct I'm quite impressed (and stand corrected)! The ones I've seen did not have any such value- I doubt that they even had 60dB of loop gain (that's the gain of the amp plus the amount of feedback).  But some of the Futterman amps have impressively low output impedance figures, such that they would have easily behaved as a voltage source, even though they made more power into higher impedances.


Kron-Hite made laboratory amplifiers in the 1960s. In their manual for the amp (which used KT88s) they claimed 80dB(!) of feedback. Its hard to imagine how they pulled that off- that amp was quite stable. I had a pair of them for a while in the late 1970s and they compared very favorably to an ARC D-75 that a friend of mine had.

This statement is false. No Futterman amp ever had anything like 60dB of feedback!! The most any had was more like 20dB, and because the amp had very wide bandwidth, oscillation (caused by its phase margin being exceeded by the feedback) was sometimes an issue. So to your closing comment here- they are well-known to **not** be stable into complex loads.

@atmasphere Roger Modjeski based his OTLs (he built several, mostly as custom projects, I have two of them) on the Futterman circuit and in his last design prior to his death he was specifically referencing the H3 circuit. In an effort to solve the oscillation issue in his first prototype and which you noted, one thing Roger found is Futterman himself omitted some things from the H3 schematic, specifically the lack of notation for the ferrite beads which Roger eventually discovered because he had me source an H3 for him to study (the NYAL versions of this circuit certainly had their issues, perhaps due to the schematic omissions as well). One look underneath and there they all were. Now when Roger completed his design I asked him how much feedback he was using. The response was 60 dB which in addition to adding the beads to the circuit (although far fewer than Futterman used) solved the oscillation issue in the second prototype of his design.

Now I don't remember why he came up with 60 dB, but since jasonbourne52 also mentioned it, I'm inclined to think it was what Futterman used. Of course this could all be a coincidence and Roger could have come up with that amount of feedback on his own through testing.

This is an interesting statement: " Tests I’ve run have shown little improvement, and this is expected given that loudspeaker systems and the drivers used therein are designed specifically with the understanding that they will be driven with a voltage amplifier."

However small, apparently there is an inprovement with current drive. Which could grow larger when speakers would be specifically designed for current drive. One can wonder why it is not used more often ... any improvement is good, even more so knowing people are willing to spend thousands on their (power, speaker, interconnect) cables alone to gain just a very minute improvement.

Is a phono cartridge a current device or a voltage device?And what about a DAC?

This is an interesting article on the subject, © 2019, Rod Elliott (ESP)
https://sound-au.com/articles/current-drive.htm#s1

One of the conclusions (scroll to the bottom of the long article):
"There have been many claims over the years that current drive is the best, and some may claim it's the only) way to drive loudspeakers, as it reduces distortion and allows the speaker to work the "way it was intended". While there is some discussion of this on the Net (see [ 2 ] as an example), there is little real evidence that the benefits are anywhere near as great as claimed. Tests I've run have shown little improvement, and this is expected given that loudspeaker systems and the drivers used therein are designed specifically with the understanding that they will be driven with a voltage amplifier. By definition, that means the output impedance is low, always below 0.2Ω, and often much less."

Julius Futterman solved the tube amp feedback problem by eliminating the output transformer. His OTL amps can have 60 db feedback with low output impedance and low harmonic/IM distortion. Yet remain 100% stabile into complex loads!

This statement is false. No Futterman amp ever had anything like 60dB of feedback!! The most any had was more like 20dB, and because the amp had very wide bandwidth, oscillation (caused by its phase margin being exceeded by the feedback) was sometimes an issue. So to your closing comment here- they are well-known to **not** be stable into complex loads.


If you doubt any of what I'm saying, just for the record I've been manufacturing OTLs longer than anyone else on the planet (over 46 years); I'm not being hyperbolic.

Because a current drive amp is only useful with a speaker that has a flat impedance. Otherwise the amp will act as a tone control and follow the dips and peaks of the speaker's impedance curve. This can hardly be called "accurate"!

How useful would a current drive amp be with a 1 ohm speaker that has a flat impedance?

PS, db Volts is calculated this way:

db = 20 x log( V original / Voltage now) 

So doubling the amplifier voltage = 6 dB louder in both voltage and sound pressure level (SPL) so long as everything is not over driven. 

Technically dB(V) is 10log(V)
Since power is V*I… and with a constant impedance is also equivalent to V^2/R…Then we use dB(W) as 20Log(V).
(I am pretty sure of it?)

Miyajima in Japan copied the Futterman circuit and made some mono amps. The late Art Dudley tested and reviewed a pair for Stereophile some years back.

There is a purity and transparency of the sound of the Futterman's that must be heard to be believed! I first heard a friend's pair driving Quad 57's back in 1978 and was instantly convinced that the Futterman OTL's are world-beater amps! 

I have three Futterman OTL's. Two mono's and a stereo chassis version that Julius made and sold as the Harvard Music H3 in the mid-Sixties. That one got a review in the old High Fidelity mag! The H3 I bought has an added pair of Altec autoformers because the first owner used it to drive a large custom-made pair of panel speakers that had ten oval cone drivers per side! Consequently these presented a low impedance that the stock H3 couldn't drive - hence the Altec autoformers!

The solution is to add an autoformer to the Futterman OTL's output. Say, 16 to 32 ohms. This will allow low impedance drive. And a corresponding increase in power/watts from 60 to 120 to 240 from a sextet of 6LF6 output tubes! 

Julius Futterman solved the tube amp feedback problem by eliminating the output transformer. His OTL amps can have 60 db feedback with low output impedance and low harmonic/IM distortion. Yet remain 100% stabile into complex loads! The drawback is that they are best used with speakers that are 8 ohms and up due to the fact that power/watts declines into low impedances - but rises with impedance increase!

Look up Ohm's Law!

Because a current drive amp is only useful with a speaker that has a flat impedance. Otherwise the amp will act as a tone control and follow the dips and peaks of the speaker's impedance curve. This can hardly be called "accurate"!

Aside from what i already wrote, here's also the practical issue. A low impedance, hgih current amp will drive anything.

This isn't quite correct, just so you know. Try putting your amp on a Sound Lab ESL sometime. That speaker is 32 Ohms in the bass and maybe about 2 Ohms at 20KHz (depending on the position of the Brilliance control). Most voltage source amps tend to sound too bright on this speaker; they struggle to make power at bass frequencies. Its MO isn't based on the Voltage Paradigm. Anytime you mix the two paradigms (Voltage and Power) you are at risk of a tonal anomaly.


SETs and other zero feedback tube amps are not meant to be used with difficult speakers (and I argue that no amp should be used with such speakers since the last thing you want to do is make any amp work hard for a living- it will make more distortion which is audible). So no amp can really work with all speakers.

Mainly because speakers are designed to be driven by a low impedance voltage source.

Aside from what i already wrote, here's also the practical issue.  A low impedance, hgih current amp will drive anything.  Others might (might) sing with speaker A and suck with speaker B. You know, like lots of high end :-)
I'm always fascinated by this. I cam plug my amp into about anything and it will work very well and sound very good. Will it produce some synergistic, euphonic magic effect? Nope.

... why are amplifiers not built as a high impedance current source?

Mainly because speakers are designed to be driven by a low impedance voltage source. When amps and speakers are co-designed though there exists the possibility to use current drive - I believe Bruno Putzeys has done this on his actives (Kiis) over at least part of the frequency band.

re: Vandersteen's being hard to drive.
I had 2Cs and then 2Cis and 2Ces.  They were the easiest, least fussy speakers to drive i ever owned.  Drive them with a 20 W NAD 3020 that could NOT drive Spica TC or SC-50s.  Drove them with a pair of tiny Lux tube amps.  They are very resistive, basically 8 ohms, and modest sensitivity (87 dB???).


Look at the impedance curves for heaven's sake.  Do they like good amps? Sure.  Do they like highly capable, high current amps? Sure, i guess. Do they need them to make music? Nope.

It would be very interesting to learn about the intricacies of the Bakoon amplifiers.They are reputed to be current drive and have high output impedance and yet seem to be able to drive many mainstream speakers without problems although they definitely sound best with genuine 8 ohm speakers rather than 4 ohm.It is surprising what speakers even the base model 15 watt will drive and they sound superb.I have owned every type of amplifier other than an OTL and the Bakoon is by far the best sounding.There is just a fundamental rightness to the sound and that is hard to unhear.
It would be interesting to know exactly how the patented SATRI circuit  works but if that information is available it is probably in Japanese.

Erik-just for clarity - I was not disagreeing with you, just adding and referencing what you said rather than blather on and being repetitive :-)