Jump to content

Revisiting "Psychoacoustics and amps" thread from '02/'03-Ken Kantor, are you there?


Guest postjob62

Recommended Posts

Pete said amplifier's clip because they run out of voltage, not out of power since according to him they never run out of current, they invaraibly run out of voltage first. So...more voltage less clipping ...according to his explanation? I'm looking forward to his response. I can hardly wait. (He's very illuminating.) :-)

Link to comment
Share on other sites

  • Replies 160
  • Created
  • Last Reply

I'm actually sitting here wondering if you are serious in this post?

You are not confused about what a Watt is, you're confused about what clipping is as I've already stated it is primarily a voltage effect. How many amplifiers have you looked at to state that 70.7 volts is the universal supply voltage? No, it absolutely is not, rather it is line distribution level in backgroung PA systems. Yes we do up the supply voltage in higher power amps, then higher voltage output devices are needed, and more of them to support the current/power bigger heat sinks, bigger power supply, etc. This is how you build a more powerful amp.

Pete B.

Link to comment
Share on other sites

Don't misquote me, this is what I said:

>For any reasonable load on an

>amp, clipping is a voltage effect because a good amp does not

>go into current limiting, rather it supplies whatever current

>is needed to provide the correct voltage until the output

>"hits the rail".

And this is what I said about amps that do current limit:

>

>An amp with a poorly designed output stage, or poorly designed

>protection might distort due to current limiting but most do

>not call this clipping.

You ought to try measuring some amplifiers on the bench rather than hit the books. I've done both.

Pete B.

Link to comment
Share on other sites

>I've built, measured, and seen more than one power amp where

>the B+ was 70.7 volts. I'm just not in an arguing mood today.

> Maybe another time.

Name those amps please so that we can verify your claim?

Here's the data off the top of my head, for amplifiers that I've scratch built, worked on, and studied over the years:

The Universal Tiger runs on +/-42V, the Citation 12, +/- 41.5, lower power versions of the Tiger amps at around +/-35V, 200W amps usually run around +/-65V, the Hafler DH500 at +/-90V and how about the Phase Linear 700 at around +/-100V.

It is your speculative and argumentative way that drives hands on professionals away from forums like this, most don't even bother. You speculate, and then proceed and draw conclusions as if said speculation were true:

>Hmm, why don't we just increase the voltage, then we'd get more >output. But don't most amplifiers use a B+ of 70.7 volts? If that >were the case, they'd all have the same output capabilities. I guess >that partly explains why they all sound the same.

Yes I'm done with this "debate" also.

Pete B.

Link to comment
Share on other sites

Soundminded-

I believe Pete B is correct here. Regardless of whether the amp runs out of current or voltage, the effect on the output signal will be the same: the output voltage will flat-top, and this is readily visible on a scope. Due to Ohm's Law, there is no condition where either voltage or current can limit invisibly, without simultaneously limiting the other.

The actual, instantaneous power consumption of the speaker is both difficult to determine and somewhat immaterial. "Power" is never the issue in clipping, voltage and/or current capacity is. Power, on the other hand, is what heats up the output stage and the speaker drivers, so it does come into play over time.

Link to comment
Share on other sites

"I believe Pete B is correct here. Regardless of whether the amp runs out of current or voltage, the effect on the output signal will be the same: the output voltage will flat-top, and this is readily visible on a scope."

The original question of how much power is needed, used, desirable for a power amplifier in a particular sound system is a very complicated one as I see it. That is why finding it is often a matter of trial and error and people tend to want to err on the high side. But how high is high enough? Yes, I agree that when an amplifier is driven into its non linear region, the output voltage will seem to hit a ceiling on an oscilloscope, in fact as the filter capacitors are depleted and the power transformer core is saturated, the power supply impedence may increase to the point where output voltage drops significantly. This does not however tell you how much power is being delivered to a load and to assume you can just calculate it by considering an 8 ohm or 4 ohm load is far too simplistic which is why I really didn't want to get into an arguement about it. There are IMO at least three factors to consider. One is how much power is actually delivered to the load. Even at a single frequncy, you need to know either the current and the phase angle between the voltage and current or the complex impedence of the load to calculate true power and reactive power. With a broad frequency spectrum electrical waveform for both current and voltage, the problem becomes at least an order of magnitude greater. Secondly, you have to know how much of that power translates into mechanical work as opposed to directly into heat. Anywhere from 95% to 99.5% is lost directly to I*I*R heating (which is one reason why I have a lot of problems understanding Vilchur's thermodynamic analysis of his AS speaker but that's another story) and then you have to know the acoustic efficiency and size of the room since the more live and smaller the room the fewer acoustical watts of output are required to achieve a given spl even at the same distance.

I am no longer of my long held and popular view that the more powerful an amplifier you have, the better. As I now see it, once sufficient power is available to achieve a desired spl without distortion, I do not see the benefit of additional power and even discounting the additional cost of larger amplifiers, there are often tradeoffs to build amplifiers bigger. Therefore, it seems to me that an amplifier with a well regulated power supply and sufficient power output capability to achieve the highest spl required from a given louspeaker in a given room is the most desirable engineering solution to the problem of how large an amplifier to use. In my particular 14 x 30 x 9.5 room of about 4000 cubic feet which is on the acoustically live side, I have found that a 60 wpc amplifier driving AR9s is sufficient. In another room also live but about half that size, I have found that an amplifier of more than twice that power driving a pair of Bose 901 is marginal to inadequate depending on the signal spectrum.

Link to comment
Share on other sites

I made this very observation earlier in the thread and was resoundingly ignored. Peak power demand in dynamic pieces can be extremely demanding and for normal listening levels and environments, is the oly justification I can think of for massive amplifiers. Other than that, many will say that a small amplifier will sound better (in the ear of the beholder) than a large amp.

Link to comment
Share on other sites

There are many possible explanations for this including the obvious ones such as the piano being played at far louder levels than a real piano or for that matter than is safe to avoid hearing damage. Another is that AR9 is particularly prone to wasting a lot of power when there are subsonic disturbances. I've got several discs which produce so much low frequency output both audible and infrasonic that the speakers actually jarred one CD player which was fairly well isolated from vibratons to the point where it knocked the laser right off the track. A subsonic filter helps conserve power for where it's needed. Unless the DBX system is properly calibrated, it can cause substantial dynamic expansion well in excess of the dynamics of real instruments. 300 watts is approaching the theoretical combined steady state capacity of the woofers and exceeds the capacity of the other drivers. I wouldn't be surprised if this runs the risk of damaging them if the signal is in a narrow enough band to be delivered to just one of them. IMO, if a greater spl is needed than a reasonably sized amplfier can deliver, the answer is to obtain more speakers and more amplifiers, not to push a single pair to its theoretical limits. Sound in excess of 100 db can cause permanent hearing loss. That was never the intent of any responsible loudspeaker designer nor the legitimate purpose of music live or reproduced.

Link to comment
Share on other sites

You perfectly describe what happens with the version of the 1812 symphony that uses real cannons... Massive subsonic frequencies demanding stupid levels of power.

Other than that, my listening room is acousticly dead beyond description and requires far to much power just to bring the music to a decent listening level. As for soundstage and positioning, forget it.

With any luck, we will be looking for a new home next year and hopefully solve that problem.

Link to comment
Share on other sites

Guest bocoogto

The example of the piano reminds me of two things I read in the 1960's about sound reproduction. Remember the Klipsch ads showing how it would take a stack of 89 "bookshelf" speakers to equal one Klipschhorn? This is based on the relative efficiencies of the two different systems.

Also, a solo piano produces peaks under normal playing conditions that, when recorded and played back, would take more than 10,000 watts of power to reproduce accurately with a typical loudspeaker system in the 90db/watt efficiency range. These are really fast transient peaks, but I think you understand the point. Whether or not the human ear could detect the lack of these fast, high level transients is unknown to me, but an interesting piece of info. Kind of makes you wonder what other instruments such as drums or xylophones may produce for transient peaks. I think it's obvious that percussion instruments alone can create such powerful transient peaks.

The premise of most of this discussion, amplifiers sounding the same, has been addressed from many angles, but not from the one above related to actual power reproduced by loudspeakers powered by consumer-type amplifiers ranging from 10 watts to several hundred watts of output power.

In spite of the above, I agree with Ken Kantor's statement that all reasonably good amplifiers sound the same. It's loudspeakers that sound drastically different from each other. The old venerable AR-3a, even compared with today's best speakers, does a remarkable job of sounding like the original program.

Link to comment
Share on other sites

The sun puts out enorrmous quantities of energy but only a tiny fraction of it reaches us here on earth. We can build a lamp as bright and intense as sunlight falling on our eyes because we only have to reproduce that fraction of the total energy which reaches us and the lamp is also much closer. However much energy it takes to reproduce a paino's full output, we only have to reproduce that fraction of it which when arriving at our ears is as intense as the sound we hear when we are in the audience, not the rest which is converted into heat without anyone hearing it. Klipsch's entire arguement was totally bogus and not in any way related to the science of sound reproduction. It was meant only to confuse those who were not on to it.

Link to comment
Share on other sites

A valid point, but this is why the piano is an appropriate choice for perspective. There are no artillery, full orchestras, synthesizers or falling drumsets in this example.

A piano's "audience" could realistically, be seated 10 feet from the instrument - maybe even at the bench, itself. And while a piano can produce an impressive spl, a significant range separates its ppp from fff capabilities. We are reasonably speaking of reproducing an acoustic instrument in its natural environment, for a near-audience experience. Such a room could even be *identical* in size & type to an audiophile's lair!

To ask a reproduction system to simulate that instrument's presence in such an environment is an appropriate and reasonable request. And this was the situation involving the DBX tape, Studer deck, and McIntosh amplifier - for the sake of argument, let's assume that the equipment was working properly, and the recording was without fault. The fact that a well-designed amplifier of substantial power ran out of gas while trying to recreate that piano, again begs the question "how much is enough?"

Link to comment
Share on other sites

A valid question, but in and of itself, its also pointless due to several intertwining factors such as.

1. What type of speaker is being used and their efficiency

2. Size of the listening room or area, and the associated acoustics

3. Type of music the listener will be listening to. Lets face it, if the listener actually likes elevator music, not much is required.

4. What the listener wants to hear.

I've listened to a 6 watt tube amp driving an extremely efficient pair of speakers, and it was respectible. No deep, earth shaking bass, but excellent for background music. This same amp would wilt driving a pair of AR9s where a massive amp is all but required to bring out its deep bass.

Link to comment
Share on other sites

Guest bocoogto

Soundminded:

Normally, your postings are logical and informative. Your comment about the energy of the sun, however, has no bearing on audio reproduction. I think I understand what point you're trying to make, but don't get the connection. It is fairly common to hear a live piano in a room the size of a typical listening room. The 10,000 watt requirement to reproduce transients created by a piano is based on that scenario. As I said, whether or not the lack of this much peak power in an audio system is a reason there would be an audible difference between "live" and recorded, is unknown.

The Klipsch advertisement was based on efficiencies of bookshelf speakers like the AR-3 at about 88 db/watt. Klipschorns are 102 db/watt+ in efficiency. I believe each 3 db increase in loudness equates to doubling the power of any sound. Doesn't this fit in the ball park of requiring 89 bookshelf speakers to equal one of theirs. Remember, we're not talking accuracy, only loudness. I think their ad was technically correct.

Link to comment
Share on other sites

As luck would have it, I have a 1927 Steinway M sized (5'-7") grand piano at one end of my music room and my AR9s at the other. Sitting halfway in between, The AR9s can generate undistorted sound at least as loud playing a piano recording with my 60 wpc amplifier as the Steinway can. This does not surprise me because the speakers can play to nearly deafening levels in that room. The way the speakers radiate acoustic energy into space is much more focused than the way the piano does. The analogy with the sun simply means that you do not have to reproduce all of the energy emitted, only the energy reaching the individual listening to or seeing the sensory stimulus. Even in a small room sitting at the piano itself or sticking your head inside the sounding board, you only hear a fraction of the energy generated. A strip of burning magnesium ribbon or an arc from an electric welder will generate a light more intense several feet away than the noonday sun. I don't know where Paul Klipsch got his numbers from. Power is a rate of delivering energy or doing work, not a total quantity of work itself. A lightning bolt has enough power to light up all of New York City...for a few nanoseconds. If a piano generates the same sound of the 0.5% efficienct cluster of AR3s at an input 10,000 watts which is 50 watts of "AUDIO POWER", that's nearly as much sound power as PeteB quoted for an entire 100 piece symphony orchestra playing fff (67 watts.) It makes no sense to me at all.

However, the question of loudness in relation to psychoacoustics brings up two important points regarding sound reproduction. One is the absolute loudness of reproduced musical instruments. I regard it as a serious distortion when instruments are reproduced too loudly or too softly. The adjustment of the volume control must inevitably be made in relation to the power of the intrument being reproduced, how loudly it is being played by the musician, and its perceived distance. Too softly and the insturment sounds feeble. To loudly and the instrument becomes an overblown caricature of the real thing. Unfortunately, only experience is a guide as to what loudness is correct.

The other has to do with the role of acoustics. It is my belief that the perceived power of an instrument is learned based on the loudness, apparant distance, and the amount and nature of reverberation. A pipe organ playing softly in a church where it may be 50 feet away and filling up the place for 3 or 4 or more seconds with each note and where there are a couple of dozen milliseconds between important initial time delays gives the impression of a very powerful sound source while a recording of a pipe organ playing much louder in your home, coming from an apparant source 10 feet away and filling up the room with reverberation lasting for 1/2 a second for each note and where the initial delays are only a few milliseconds apart gives the impression of a much smaller source of acoustic power. The difference also has to do with the energy under the time/intensity curve associated with each note which is a non linear relationship that enhances musical dynamics. This is one more reason why IMO it is not possible to accurately reproduce music without reproducing the acoustical effects of the normal venue that music would be heard in. It may be relatively unimportant for reproducing a small jazz combo normally heard in a nightclub but for reproducing the sound of the Mormon Tabernacle Choir as it would be heard at the Tabernacle or the sound of the Boston Symphony Orchestra as it would be heard at Boston Symphony Hall, the difference is immediately apparant to anyone with normal hearing.

Link to comment
Share on other sites

A good point!

When speaking of recording unamplified instruments in performance, and then accurately reproducing that performance, it's probably safe to say that the capture of a performance venue's acoustic effect upon the music is as an important a factor in the successful reproduction of the original event as any other.

This is a difficult nut to crack, though. If the reproduction system is to be capable of successfully simulating the original event, wouldn't it have to "know" what the original sound was like, and then adapt/modify its output into the specific room in which it finds itself?

In other words, shouldn't the reproduction system have as pure a set of data as possible (unmodified, and unadulterated by EQ, compression, effects, etc.) to work with, and then to modify in whatever way necessary to make the listening room/experience a simulacrum of the original?

Link to comment
Share on other sites

"This is a difficult nut to crack"

You don't know the half of it. I've been at it for over 31 years.

"When speaking of recording unamplified instruments in performance, and then accurately reproducing that performance, it's probably safe to say that the capture of a performance venue's acoustic effect upon the music is as an important a factor in the successful reproduction of the original event as any other."

Unfortunately it is not possible to "capture" the performance in this way in the sense of recording it if for no other reason than the inability to isolate the reverberant sound field from the direct field. The microphones which are intended to pick up just the reflections invariably pick up the sound coming directly from the musicians as well and that gets reproduced by speakers which are only supposed to capture the reflections. Among many flaws in quadraphonic sound of the 1970s, this was one I found especially objectionable. Perhaps someone will invent a way to overcome this but so far, I haven't heard of one. Therefore the only practical way is to regenerate the reverberant sound field by devising an arrangement of electronic circuits and an array of loudspeakers which produces a comparable effect.

"If the reproduction system is to be capable of successfully simulating the original event, wouldn't it have to "know" what the original sound was like, and then adapt/modify its output into the specific room in which it finds itself?"

For absolute perfection in recreating a particular seat in a particular venue, you'd have to have data on the exact relatinship between each instrument on the performing stage and the listener's seat. The relationship for every pair of directed coordinates is different, even in the same concert hall. What you'd hear in any two seats would be different and if the musicians moves to another spot or even reorients his instrument that also has the effect of changing the relationship.

"In other words, shouldn't the reproduction system have as pure a set of data as possible (unmodified, and unadulterated by EQ, compression, effects, etc.) to work with"

Again, for absolute perfect simulation, you'd have to start with a recording which is as perfect as it can be made having flat frequency response, no compression, and no recorded echoes or reverberation. Not only doesn't this exist for real recordings we have today and not only can't you precisely simulate the very complex relationship as accurately as you like but the acoustics of the room you listen to the playback in superimposes its own acoustical effects on the sound.

The good news is that reasonable facsimiles of acoustical effects of concert halls for many recordings are possible although difficult to "tweak" and almost any credible recreation is far better than the type of sound we hear without one at all. At least that's my opinion.

Link to comment
Share on other sites

My suggestion was not to calculate how much power is needed, but rather to determine if your amp, on your system, at your listening levels is clipping. Your addressing a different question and your making it overly complicated. Therefore, I will drop out of this discussion.

I'd be surprised if your Dynaco amp did not clip seeing as it only supplied 40 W into 4 ohm loads. Your new amp probably does much better and might not clip if you listen at lower levels than most. Another consideration is that I do not set up my serious system for compressed source material which might not clip under most circumstances, rather I search out material with less/no compression and try to set up the system to handle the best source material.

Your comments seem to be aimed at disproving Ludwig's claims, yet he backs them up with calculations and measurements and they roughly agree with what I've known for many years. I've not checked his calculations since my amps have clip lights so I know when my amp clips, and it does especially on source material with wide dynamic range, or without compression when I run only one amp.

I should point out that there are other ways of observing clipping. One can run the scope in X/Y mode with gain correction and view a straight line when in the linear region, then see the well known bend at the extremes when the output does not follow the input. This is well known, and I did it as a kid.

I find it interesting that people on these public forums think they can figure things out on the spot, or that they need to, when the reality is that it's probably already been figured out by much more competent people often in a peer reviewed environment. I used to cite the papers more, and I did on other subjects in this thread, however they are often ignored with the discussion continuing with far fetched speculation and ideas. This is probably why most well read professionals don't bother with these forums. All of this is fine for casual conversation but one must be very careful when making claims of fact.

Experiments must be controlled. I don't mind casual observations and conversation about audio but I don't take them necessarily as fact.

Pete B.

>"I believe Pete B is correct here. Regardless of whether the

>amp runs out of current or voltage, the effect on the output

>signal will be the same: the output voltage will flat-top, and

>this is readily visible on a scope."

>

>The original question of how much power is needed, used,

>desirable for a power amplifier in a particular sound system

>is a very complicated one as I see it. That is why finding it

>is often a matter of trial and error and people tend to want

>to err on the high side. But how high is high enough? Yes, I

>agree that when an amplifier is driven into its non linear

>region, the output voltage will seem to hit a ceiling on an

>oscilloscope, in fact as the filter capacitors are depleted

>and the power transformer core is saturated, the power supply

>impedence may increase to the point where output voltage drops

>significantly. This does not however tell you how much power

>is being delivered to a load and to assume you can just

>calculate it by considering an 8 ohm or 4 ohm load is far too

>simplistic which is why I really didn't want to get into an

>arguement about it. There are IMO at least three factors to

>consider. One is how much power is actually delivered to the

>load. Even at a single frequncy, you need to know either the

>current and the phase angle between the voltage and current or

>the complex impedence of the load to calculate true power and

>reactive power. With a broad frequency spectrum electrical

>waveform for both current and voltage, the problem becomes at

>least an order of magnitude greater. Secondly, you have to

>know how much of that power translates into mechanical work as

>opposed to directly into heat. Anywhere from 95% to 99.5% is

>lost directly to I*I*R heating (which is one reason why I have

>a lot of problems understanding Vilchur's thermodynamic

>analysis of his AS speaker but that's another story) and then

>you have to know the acoustic efficiency and size of the room

>since the more live and smaller the room the fewer acoustical

>watts of output are required to achieve a given spl even at

>the same distance.

>

>I am no longer of my long held and popular view that the more

>powerful an amplifier you have, the better. As I now see it,

>once sufficient power is available to achieve a desired spl

>without distortion, I do not see the benefit of additional

>power and even discounting the additional cost of larger

>amplifiers, there are often tradeoffs to build amplifiers

>bigger. Therefore, it seems to me that an amplifier with a

>well regulated power supply and sufficient power output

>capability to achieve the highest spl required from a given

>louspeaker in a given room is the most desirable engineering

>solution to the problem of how large an amplifier to use. In

>my particular 14 x 30 x 9.5 room of about 4000 cubic feet

>which is on the acoustically live side, I have found that a 60

>wpc amplifier driving AR9s is sufficient. In another room

>also live but about half that size, I have found that an

>amplifier of more than twice that power driving a pair of Bose

>901 is marginal to inadequate depending on the signal

>spectrum.

Link to comment
Share on other sites

I agree Ken, but I don't think about the current limited case much since I usually try to set up a system with an amp that does not go into current limiting. Obviously, it can happen and the voltage might flat top, or as with the reports of erratic output protection circuits it sometimes oscillates. I've not observed this but there have been plenty of reports.

Pete B.

>Soundminded-

>

>I believe Pete B is correct here. Regardless of whether the

>amp runs out of current or voltage, the effect on the output

>signal will be the same: the output voltage will flat-top, and

>this is readily visible on a scope. Due to Ohm's Law, there

>is no condition where either voltage or current can limit

>invisibly, without simultaneously limiting the other.

>

>The actual, instantaneous power consumption of the speaker is

>both difficult to determine and somewhat immaterial. "Power"

>is never the issue in clipping, voltage and/or current

>capacity is. Power, on the other hand, is what heats up the

>output stage and the speaker drivers, so it does come into

>play over time.

>

>

>

Link to comment
Share on other sites

  • 1 month later...

From "An Ultra-Linear Amplifier" by David Hafler and Herbert I. Keroes, published in Audio Engineering, November 1951:

"Another factor of considerable importance in evaluating amplifier performance cannot be seen from the curves. This is overload characteristic. The amplifier has been given listening tests under overload conditions with a pad on the output so as not to deafen the participants. Peaks which would require a 40 watt amplifer are transmitted without irritation even though the output can be seen to clip on the 'scope. The overload recovery is rapid and has no noticeable hangover, so a clipped peak has no time to penetrate the ear. Some amplifiers break up on a peak, and for seconds thereafter the sound is distorted badly because of poor recovery. In the ultra-linear amplifier transient instability has been eliminated--changesin amplifier characteristics caused by overload do not make the circuit unstable; and, therefore, recovery is almost instantaneou. Most feedback amplifiers fail miserably under overload listening tests."

Sounds a bit like advertising hype but there is probably much truth to what they write, even today.

Pete B.

Link to comment
Share on other sites

  • 1 month later...

Ed, I think you asked about power supply capacitor size somewhere in this thread.

I posted this about power supply cap size on another forum:

A power amp with unregulated supplies for example, lets say something like 50W/ch, will idle with about +/- 40 V on the power supply. If you load it with a dummy load and drive it to clipping on a bench, the supply will sag typically to about +/- 30V for this type of amp and more 60/50 Hz ripple will be present.

Now remember that power = V^2/R so if the supply stayed at +/-40 it would be able to deliver much more power, and indeed this is where the dynamic headroom comes from. The supply does not sag much during short peaks in the music and the amp can deliver more power during these peaks. Where there are more repetitive peaks of longer duration the supply sags, bigger caps provide a longer time constant so that more repetitive peaks can be handled without clipping and there will be less 60/50 Hz ripple.

It really does make a difference and even many years ago Dynaco sold an add on capacitor box for the ST400, I think it was 100,000 uF.

It is always interesting to think of things at the limit, if we think about infinite caps the supply would never sag and then the RMS continuous rating of the amp then becomes equal to the dynamic headroom value, it would also have no ripple. This is similar to a regulated supply.

Here are some quick calculations for a typical 50W example, assuming that the output stage comes within 2 Volts of the rails and does not current limit:

Idle Full power 8 ohm Full Power 4 ohm

V+/- 40 V 30 V, 27 V

Vpeak out 38 28, 25 = V+/- - 2

Vrms out 26.9 19.8, 17.7 = Vpeak out*.707

8 ohm power 90 W 49 W, N/A = Vrms out^2/8

4 ohm power 180 W N/A 78 W = Vrms out^2/4

The power at idle is what the amp will put out for short peaks without clipping and so we can compute the dynamic headroom:

8 ohm dynamic headroom = 10 log (90/49) = 2.64 dB

4 ohm dynamic headroom = 10 log (180/78) = 3.6 dB

Pete B.

>I've been searching (to some extent in vain) for information

>as to which, if any vintage Marantz SS receiver might do a

>better job than my current Pioneer SX-750 in powering my

>2ax's. The Pioneer does a great job, but my Marantz 2230

>sounds *so* good with my 4x's that I just thought there might

>also be a magical Marantz out there for my 2ax's.

>

>In so doing, I came across the above referenced thread in

>which Ken Kantor basically put forth that which he considers

>irrefutable, to quote:

>

>"If two solid state amps of comparible power sound different,

>then one or both of the amps is broken. Properly functioning

>amps do not sound appreciably different from one another, at

>least different enough to change the sound quality in a system

>to any real degree."

>

>I continued to read the thread, and if I'm correctly

>interpreting Ken's scholarly statement, a solid state watt is

>a solid state watt, just that and no more; further, if two

>completely different amps of exactly the same power output are

>run through the identically same speakers under identical

>circumstances, then the results are sonically

>indistinguishable one from the other, at least to the human

>ear.

>

>Say what?

>

>Now Ken is obviously a brilliant engineer and I'm an

>electrical unsophisticate, so unlike others I'm not here to

>refute his point(I couldn't) but rather to ask Ken and any

>others to amplify (no pun intended) and clarify, perhaps in

>more layman-like terms. I would have to assume that whatever

>the chosen power level, it would have to remain constant for

>whatever impedance the speaker presented to the amp. Obviously

>you couldn't use two different amps producing 100 wpc at 8

>ohms with 3a's if one of the amps folded up with a 4 ohm load.

>But beyond that, am I to understand that *nothing* else about

>the amp matters? Output coupling, power supply, transformer

>size, damping factor, subjective "voicing", and all the things

>that make up a SS amp's architecture-none of these things

>matter, just the number of watts produced?? So the perceived

>"warmth" of an early Marantz, the smoothness of the HKx30

>series, the brightness and articulation of '70's Pioneers- all

>this is simply that, just subjective perceptions based on,

>well, perceptions?

>

>I've always held forth that perceptions are more important

>than reality, and in fact often become reality. So maybe this

>is my just desserts. But honestly, I just had a deja vue

>recollection to the exact day I found out there was no Santa

>Claus. And I'm 56!

>

>Admittedly, I'm of a mindset to want to believe this. I

>recently read Peter Aczel's "10 Biggest Lies" article, and

>frankly, a lot of what he said made sense to me. But I need

>some help with this one. Ken, if you're there, please make

>sure I understand this correctly. It will save me a lot of

>agonizing over non-existent details, and will free up some

>time to concentrate on what I know does matter- the wonderful

>sound of these old AR speakers.

>

>As an addendum, Ken, how do you and others weigh in on the

>Aczel "tube superiority is a myth" proposition?

>

>Regards,

>

>Ed

>

>

>

>

Link to comment
Share on other sites

No mystery here, just straight forward electrical engineering. The larger capacitors store more charge the depletion of which is directly related to the drop in power supply voltage. Short demands for power deplete relatively little charge and therefore not much drop in supply voltage allowing them to recover quickly whereas sustained "full power" demand will eventually run the caps on "empty" and straight off the transformer. This of course translates to an increase in effective power supply impedence equivalent to the transformer inductance, hence, the larger the power transformer, the more available power assuming the diodes and output devices can handle it. The availability of high quality high power audio amplifiers at modest cost should make it possible for audiophiles even on modest budgets to provide all of the power they would ever need to stay out of the trouble zone. At least one designer went to the absurd lengths of using strings of automobile batteries, an exceptionally foolish and dangerous idea IMO.

As for Ken's statement that all solid state amplifiers sound alike, IMO that is nearly true but not quite 100% correct since minor differences in overall system frequency response which includes differences inherent in amplifiers themselves and difference when operating into real world loudspeaker loads can be audible with careful listening but are of relatively minor importance. I think this reflects Julian Hirsch's view as well. As for vacuum tube amplifiers, IMO they are usually poor performers by comparison and some of the latest and most expensive low power SET amps plain stink. They have awful frequency response measurements and distortion orders of magnitude greater than well designed solid state amplifiers (which may still be below the threshold of audibility however.) They also suffer from poor control over loudspeakers owing to their high output impedence as reflected in their low damping factors and truely awful hysterisis and eddy current losses in their output transformers. The one exception is OTL amps such as the Futterman/NYAL design which seemed to me to perform comparably to solid state amplifiers. Whether they are worth the cost and trouble is another matter. Tube amps may be useful in producing better sounding systems with many modern loudspeakers because their complimentary shortcomings of rolled off treble and boomy bass compensate for the shrill thin sounding loudspeakers audiophiles mistakenly gravitate to, often being inexperienced listeners auditioning them in rapid fire AB demos in acoustically dead showrooms. It seems to me, once they get them home they will go to any lengths to mitigate their mistake from buying tube amps, vinyl records played with mc cartridges and high capacitance speaker wires and interconnects rather than replace them. They say they want accuracy but I don't think most of them know what that is.

Link to comment
Share on other sites

  • 2 weeks later...

Most of us probably like to think that these products, especially high end designs, are designed by people in white lab coats who could design part of a NASA space electronics system. But as time goes on and I look at one design after another, deficiencies even mistakes, or at least poor design practice can be seen in most things.

I'd seen the ads and reviews of Bryston, impressive, then I read a review in the Audio Critic which I thought was a top notch magazine. It's not bad, but the publishers biases came out over time. I found it odd, very odd, that he claimed to be so scientific yet dismissed the Hafler Null test when reviewing the XL-280, strange since a null test especially the way Hafler did it is proof of real world performance.

I read a Bryston review and it tested with outstanding performance, there were comments from their hired PHD electrical engineer, a professor I believe, that raved about what outstanding engineers designed the Bryston. I believed all this, never interested enough to look at the Bryston schematics.

When I finally did I saw roots in the South West Technical Products Tiger .01 amplifier from the early 1970s. Some from that amp, and some from their bigger amp, I know these designs very well and as an engineer it's easy to see these things in the schematics. These amplifiers were known to have serious problems.

Here are more of my comments about the Bryston:

http://www.diyaudio.com/forums/showthread....1605#post531605

http://www.diyaudio.com/forums/showthread....8425#post538425

A gentleman in this thread has a Krell amplifier where one of the boards "set itself on fire" as he put it:

http://www.diyaudio.com/forums/showthread....42357&highlight

He reverse engineers it, finds transistors that are running over their rated power, corrects this adds a few other improvements and designs a plug compatible circuit board to rebuild it. Nice work. Shows that designers don't even calculate Pdiss in every active component. I like Krell, more or less.

Pete B.

>I had a 4B about 10 years or so ago when I bought the

>Paradigm Studios. I had a Classe preamp which was noisy (hiss)

>by comparison to the other preamps I had. Anyway I accidently

>pulled the interconnects from the pre to the Bryston out. I

>heard a loud crack and that was it. When I reconnected and

>tried to listen to music there was no bass, The Bryston fused

>the voice coils on all 4 woofers. Fortunately the dealer and

>Paradigm were very accomodating and replaced the woofers under

>warranty. These were no lighytweights... they easily handled

>the power of the 4b to nose bleed sound levels.

>

>Back to the 4B. It was obviously very clean to any level and

>had no problem driving the 6ohm Paradigms. Compared to

>subsequent power amps I used on those speakers the 4B's had a

>deeper more powerful bass. I know it didn't actually go deeper

>but the interaction with those speakers was such that it

>sounded deeper and more powerful and bass was very tight and

>clean. Amazing amp but use speaker fuses...!!!!

>

>I had a 2B prior to that. It was also excellent but nothing I

>have had since matched the bass and subjective power of the

>Bryston. I had a Marantz 510M and it was nice but either

>Bryston in my opinion sounded more powerful. This is

>subjective because I did not have the Marantz at the same time

>as the Brystons.

>

>It was wonderful.... should have kept that sucker.

>

>

>

>>I'm curious as to what people here think of Bryston power

>>amplifiers?

>>Yes this is a trick question, but please go ahead and

>>comment.

>>

>>I had a very good opinion based on reviews and comments from

>>several trusted friends, I've probably even heard them

>>somewhere along the way.

>>

>>Pete B.

>

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...