Jump to content

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


Guest postjob62

Recommended Posts

I've heard so many audiphiles make so many rediculous blanket statements about tubes and transistors, I couldn't resist. Here are some of the more common ones; "Tubes sound warm", "Tubes sound musical" "Transistors sound harsh and grainy" "Transistors produce harmonic distortion with even harmonics which are unmusical while tubes produce harmonic distortion with odd harmonics which are musically related." "The soft clipping of tube amplifiers is more musical than the hard clipping of transistors." The list goes on and on, we've all read and heard them. You can find lots of awful tube amps which sound harsh. You can find lots of tube amps which far from having a "bloated" bass have little or no bass at all. You can find tube amps which have rock solid bass like McIntosh MC3000. You can find transistor amps which are fuzzy as all hell like an early model called Acoustech which comes to mind. Or those with muffled highs like the one ElectroVoice sold which had a badly rolled off high end like many tube amplifiers. I find it impossible to generalize except to say that the best transistor amplifiers sound clearer than most tube amplifiers which IMO is one reason tubes all but died out by the 1970s. When I heard the NY Audio Labs OTL tube amplifier in 1983, it sounded to me exactly like a solid state amplifier, every bit as clear and accurate. I think Ken Kantor was right in this sense, the audible variation among vacuum tube amplifiers caused by their limitation is far greater than among solid state amplifiers. It is very inconvenient for an industry which depends on a constant parade of new and improved models to thrive when relatively inexpensive long existing technology offers little or no disadvantage over the latest, greatest, and most expensive. It is therefore the well educated consumer's boon that this vast reservoir of very low cost used equipment is out there virtually for the taking. And it would hardly surprise me at all if the 85wpc Pioneer SX950 receiver I bought at a garage sale for $10 with a slight fiddling of its frequency response couldn't be made ot sound exactly like a very expensive separate power amp and preamp to most people under most circumstances, certainly all but those which challlenge it to exceed its design limits.

Link to comment
Share on other sites

  • Replies 160
  • Created
  • Last Reply
Guest bocoogto

Good posting, Soundminded! You confirmed some key points about tube and transistor amps. I also believe the good tube designs from McIntosh and Dynaco and others sounded very much alike--and also similar to well-designed transistor amps. Some of the early transistor amps like the Phase Linear 400 and Acoustech had serious design flaws. I once tested the Phase Linear 400 with my AR3's and lost two tweeters in the process. Too much power or poor design of the amplifier??

Yes, the $10 Pioneer receiver can be tweaked to sound like a more expensive preamp/amp pair. I've done this, although in a much less controlled test than magazines like Hi-Fi & Stereo Review did in the '80's. You would never see such an effort by Stereophile magazine, in which sound quality is always directly related to dollars!! If it was possible and you built a PERFECT amplifier that had zero distortion and huge amounts of power that could be built for $100, Stereophile would hate it. I remember reading an article in that magazine about the shortcomings of A/B testing. They concluded that long-term listening was more accurate than blind A/B tests. What a crock!! I'm sure you all have listened to your system one day and changed nothing, but the next day you hear differences that aren't there.

Link to comment
Share on other sites

Stereophile Magazine is IMO the absolute worst consumer magazine in any product line I know of. It seems more like an industry advertising blurb than an objective evaluation of products. For the longest time, they didn't even publish any measurements at all and I don't know that they do much of that even now. It is geared towards the young wanabee male who has more money than knowledge or experience to help sell vastly overpriced products. Look at other message boards and you will see most "audiophiles" have traded so much equipment in so short a time and lost so much money at it that it makes my head spin. And the biggest joke of it is that they never learn, they just keep on doing it. As for the manufacturers, kachung, kachung, kachung, the sales and profits just keep ringing up. What a racket.

Link to comment
Share on other sites

Pete,

would you mind attaching the diagram for the Hafler Null test or sending a current link to the schematic.None of the links I tried worked or they needed a password.

Got moved and have set up some of my hifi stuff. Comparing amps and speakers. Amps Crown DC300A, Hafler DH200, Kenwood M1 etc.

Speakers, AR12, Paradigm studio monitors, ARLST, Large Advents, Dynaco A25XL, etc.

This is just for fun, but I am interested in the principle and application of this Null Testing.

I am not a believer in exotic cables, special feet, SET amps... but I believe I hear differences between preamps, power amps and for sure speakers. Never heard any differences between speaker cables or interconnects. I suspect that the differences I hear in the electronic components are due to speaker interaction and to possible component ageing etc.

A few months back I modified a couple of Crown preamps. They are generally regarded as harsh and bright by the 'hifi' crowd. I believe I have made my 2 samples more 'listenable' but would like to do something like the null testing to scientifically validate my impressions by comparing to a 3rd unmodified IC150 preamp and to compare to my Haflers, McIntosh and APT Holman preamps.

thanks and regards, Wally

ps still haven't unpacked my test gear but will get to the measurements on the LST autoformers as soon as I do.

Hey Bret... am hooking up the LST's tonite to see what they sound like in my new place. Small room so I'm not sure I can do them justice here.

>I agree with Ken here.

>

>Bret, do you actually think that those doing the research have

>not thought of trained or golden ear listeners? The golden

>ears claim is a very common come back from those who think

>audio listening is comparable to wine tasting. Some say that

>Dyslexics have more acute hearing, are you Dyslexic? Not that

>better hearing would make a difference with good amplifiers.

>

>Many try to offend the technical types claiming they do not

>have good ears.

>

>It's interesting that I've worked in super computer design,

>radar and communications systems, digital design and more, and

>there is never some magic unknown "thing" that causes

>problems. These areas are all significanly more complex and

>more difficult than audio amplifier design.

>

>Those talking as if they're authorities on amplifier design or

>the "sound of amplifiers" thinking there is more to it than

>loaded transfer function, nonlinearities of all types, and

>freedom from oscillation and current limiting ought to become

>versed in the work of Baxandall, the Hafler null test, and

>more. Tube and single ended amps sound different because they

>often have highish output impedance, and/or high levels of

>distortion.

>

>I discuss the Hafler null test here, I'm PB2:

>

>http://www.diyaudio.com/forums/showthread.php?threadid=48873

Link to comment
Share on other sites

> the audible variation among vacuum tube amplifiers caused by their limitation is far greater than among solid state amplifiers<

I would be a liar if I said that I have enough experience of this to be able to generalize from personal experience. I have every reason to suspect it is true.

You know what's great about tube equipment, though? You can just keep changing tubes until you get it to sound like you want it to. I have no idea why anyone wants to buy a $12,000 tube amplifier and spend another $2,000 on tubes to change-out the originals in order to improve the amplifier. Odd behavior if you ask me.

The only reason I bought tubes for the Jolida was that I couldn't believe how bad the analog outs sounded compared to the Perpetual Technologies gear connected to the digital out. One set of tubes did improve it quite a bit. I'm not talking about "tubes as an equalizer," either. The stock tubes really stank and made the Jolida unlistenable after comparing it to the PT gear. Even after settling on a pair of tubes that sound pretty good, I still prefer to listen to the PT gear. In fact, connecting the PT gear to a Toshiba DVD player I got "free" with the purchase of a TV set (the player retailed for $89) resulted in a vastly better source than the stock Jolida.

Isn't dis-ing a Phase 400 like walking under a ladder, breaking a mirror, or spilling salt? You might want to knock on a wood side-panel or a genuine American Walnut veneer cabinet.

Did anyone ever do any tweaking of Phase 400s? I'd love to have a home theater run off a bank of Flame Linear 400s with all the meter lights working. And then I'd need a 7.1 DVD-A of Whole Lotta Love, a couple of color organs, and some black-light posters.

Far out, man.

Dynaco amps - If the tubes were so superior you have to wonder why David Hafler abandoned them. I really hate that he's no longer around.

Bret

Link to comment
Share on other sites

I must say I am amused and amazed at the seemingly infinite number of variants the human mind can invent for what is inherently a simple idea with only a relatively small number of basic approaches. The British term "valve" sums it up for me pretty nicely, a power supply and what amounts to a rheostat to control the flow of current to a load. What are there, perhaps over 10,000 different amplifiers which have appeared on the consumer market over the last say 50 years or so, the vast majority of them being solid state class AB amplifiers, a smaller number of vacuum tube counterparts and a smattering of class A types. A few oddballs here and there like OTLs and switching amp here and there and that's about it (did I omit anything significant except maybe how much feedback to use?) And every one of them claiming to be superior to every other one, at least in its price category by virtue of some magic bullet the designer alone had the insight to discover and exploit. I'm surprised that by this time, some clever inexpensive digital amplifier hasn't come along to make all of the analog units as extinct as the dinosaur and replace them by being virtually textbook perfect. I expect that to come along in the next few years and for audiophiles to say as usual that they don't like it. For those of us still in the real world, that's what we'll buy when we need a new amplifier, 200 wpc, indestructable, no distortion, no noise, no deviation from flat response and only $100. Sayonara Krell. Adios Mark Levinson. Hasta Le Bye-Bye Bryston.

Link to comment
Share on other sites

I'm in Europe so will be brief:

There is no analogy between a capacitor and an amp, in my mind. They do not have congruent functions. Plus, mylar capacitors have totally different published specs than electrolytics. It seems obvious to me that changing a cap to one that has a lower ESR could change the sound in a known and predicatable way. I don't see what that has to do with the relative audibility of amps. ??

Link to comment
Share on other sites

Given the inherent limitations of a DAC, I'll be surprised if they ever manage to build a digital system that sounds as fluid as an analog system. Our ears are analog after all and no amount of engineering will ever change that, and they unfortunately (coupled with our brains) can hear far more detail than we can easily explain. Consider the fact that with only two ears, we can identify whether a noise comes from in front or in back of us.

The Denon DCM 560 CD player was even marketed as being almost as good as analog.

Link to comment
Share on other sites

Hi Wally,

Here are some links.

The null test is shown on page 7:

http://www.hafler.com/techsupport/pdf/XL-280_amp_man.pdf

This references some earlier work by Baxandall:

http://sound.westhost.com/sim.htm

Good to hear that your in your new place.

Pete B.

Link to comment
Share on other sites

Hi Soundminded,

You've mentioned your Dynaco Stereo 120 several times here. I'm just wondering how familiar you are with the circuit topology, do you notice any advantages or disadvantages in the design?

I studied this design when it came out many years ago and compared it to the HK Citation 12, and the SWTPC Universal Tiger.

Pete B.

Link to comment
Share on other sites

Hi Pete, haven't heard from you in a long time. The Dynaco Stereo 120 had one huge advantage, the bias was fixed in the design, there were no pots to adjust ever. It was also pretty well protected. I once accidentally shorted the B+ to ground and sparks flew. It still worked perfectly for another 20 years after that. It had a regulated power supply. That was the main difference between it and the Stereo 80, the 80s lack of regulation. It was also a very simple easily reproducible design as were all Dynaco circuits. My speaker arrangement has about a one ohm impedence at high frequencies by virtue of the three 8 ohm tweeters in parallel with the AR9. It never seemed to bother it. It finally died after 23 years of valliant service. Its left channel was taken out by Alexander Nevsky's battle on the ice. It was a casualty of war. I replaced the output transistors but they failed again and I decided to build a new kit which I've been very pleased with, the Mosfet 120 is no longer available sadly. I should have bought several of them. One my Dynaco SCA80Qs failed recently too. I haven't had a chance to look at it but I think it's got a power supply problem. There are some 5 watt resistors which have had a spotty record. I also had to replace one of the large electrolytics. Turned out Mallory makes exact replacements cheap.

Any progress on your AR9 based speaker concept?

Link to comment
Share on other sites

>I don't see what that has to do with the relative audibility of amps. ??<

Amps are made-up of parts. Some parts are different than others. Some specs (damping, current, bias) are different than others. There are even capacitors in there. Thus I continue to wonder how lowering one cap's ESR in a speaker changes it from another speaker, but changing damping or bias or a capacitor (even power supply caps)in an amp does not change it in an audible way.

Why does anyone bother with huge power supplies if power supply differences make no audible difference?

How can changing one capacitor in a speaker change it, but changing resistor types across the outputs of an amplifier, not? Surely just *that* subtle change changes something.

Bret

Link to comment
Share on other sites

It's really very simple: the difference all comes down to whether the final signal is impacted in a way that is audibly detectable. In the case of the crossover caps, the signal passes through them, and they are used in a way where the expected design deviations can reasonably be exected to lead to audible changes to the signal. (low impedance with no feedback, etc.)

In the case of an amp, on the other hand, a good topology is fairly insensitive to this kind of component change... it just doesn't yield a significant impact on the actual signal. This is why it is almost impossible to distinguish between amps. This last assertion is a widely proven fact, not a speculation on my part.

-k

(3am in Jutland.... I'm going back to sleep.)

Link to comment
Share on other sites

>In the case of an amp, on the other hand, a good topology is fairly insensitive to this kind of component change... it just doesn't yield a significant impact on the actual signal<

Okay.

So. . . uh, why the move to DC (direct coupled) amplifiers to remove the capacitors from the circuit?

Nevermind, treat it as rhetorical.

You win. All amplifiers sound alike. Next time someone asks I'll tell them any amp will do.

Link to comment
Share on other sites

I'm going to quote myself from earlier in this thread:

"Many types of studies have been conducted over the years, and the differences between even the best and worst amps are surprisingly small. This does not mean that they are never audible. It just means that the actual sonic difference between a 50W receiver from 1975, and a modern $5,000 'Class A' rave, is not at all what people have come to accept. Believe me, if real differences could be demonstrated, companies would be all over it."

Personally, I don't think this is a field of discussion that is well served by cynical irony, or hyperbole. I'm just relating facts as I, as an audio professional, understand them. I am certainly not at all interested in "winning" any arguments with you, nor am I likely to be shamed by wit into ignoring what I have learned over the years.

Link to comment
Share on other sites

>(3am in Jutland.... I'm going back to sleep.)

Wow...89 years later, and this Forum is fighting the new Battle of Jutland. (You historians out there will understand my reference.)

This new version of the battle is at least as epic, and, apparently, at least as inconclusive to some.

Steve F.

Link to comment
Share on other sites

Of course a harp is not digital. (Well, it IS played with digits...)

The basic function of a basilar membrane is (mostly) analog. The nerve cells connected to it are pulse encoded, and the communication of sound vibrations to the brain is completely digital in nature. Here's a description from a text on auditory physiology:

"Like in all other sensory systems, information about the outside world is carried to the brain in trains of all-or-none action potentials in ensembles of peripheral afferent nerve fibers. As applied to the auditory system, the term "code" is simply a way of describing the manner in which information about sound is represented in such neural activity."

Link to comment
Share on other sites

Each string of a harp produces one note, each cilia in the inner ear detects one tone or note. The activeation of multiples activates what we hear.

Digital is nothing more than a series of zeros and ones, no more. Digital music as we all know attempts to recreate a sine wave with a series of incredibly small digital steps resulting in a stair step shaped sine wave. DACs have come an incredibly long way since the first CD player and do a far superior job now versus then, but at their heart, this is what they still do.

IMHO, to say the inner ear is "digital" is a gross over simplification and an improper comparison, one that lends itself to arguing that the harp is digital given the distant similarities between its strings and the ears cilia (small hairs in the cochlea.

Link to comment
Share on other sites

Since the topic of this thread is psychoacoustics, a branch of psychology, I found it interesting that in the course and text I studied about psychology when I was in school, so much of psychology has to do with the measurement of sensory perception. Human hearing like the other four senses has a limit to its ability to resolve small differences as well as a limit in its frequency and loudness range. Let's say that the smallest increment a human can hear is 1/4 decibel difference (human hearing is probably no better than half that or 1/2 db.) The range of loudness of the human ear is 120 decibels which stretches from the threshold of hearing at the softest end to the threshold of pain at the loudest. So to cover every possible level of loudness a human could hear would take 480 different possible loudness levels. But even the widest dynamic range of music, certain classical music has a range of no more than about 73 decibels which is the noise threshold specified by AIA for concert halls (27db) to 100 db the loudest crecendo so to reproduce all levels of classical music would require only 292 different loudness levels. Even a modest cd player is capable of over 90 db of range. The 16 bit word length of the RBCD standard allows 2 to the 16th power of different levels or over 64,000 different loudness levels and it can change from any level to any other level up to 44,000 times a second on each of the two channels. This gives it a bandwidth of over 20 khz. This exceeds the ability of the human ear/brain to resolve differences and covers the entire range of human hearing. The digital end of the recording/retrieval system in RBCD is virtually flawless unless the recording is damaged. Therefore the only limitations are in the linearity of the A/D and D/A converters which continue to get better, even the very cheap ones. BTW, this is entirely consistant with informantion theory which defines the equivalent analog channel height and width required for sound transmission or storage and retrieval as well.

The test of a digital system is not whether a digital reissue of an old analog recording sounds the same, better than, or worst than the analog equivalent because it was made at a different time, the source tape being in a different usually deteriorated condition and using entirely different editing methods but whether a digital recording made directly from an analog source sounds identical. If you have a cd recorder, why not try dubbing some vinyls and see what you get. Even today's cheap units do a very credible job, far better than the best analog cassette recorder ever could. Comparing them A/B is easy becuse it is so easy to jog the digital playback to synch it exactly with the analog recording.

Link to comment
Share on other sites

This is correct. Muscle fiber contraction and neuron firing work in basically the same way. Once the action potential threshold is exceeded, Na+ ions enter the cell and K+ ions leave through the cell membrane. Each one fires exactly the same way. After firing, the sodium/potassium ion pump restores the original difference in ion concentration by pumping sodium ions out and potassium ions in. This is not an analog process. Electrically across the membrane, the firing looks exactly like a digital pulse and is due to the difference in the electropositivity of Na+ and K+ ions. Ca++ ions can block the channels in the membrane through which the ions exchange quieting the nerve by raising the required stimiulus to initiate firing.

Link to comment
Share on other sites

Archived

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


×
×
  • Create New...