I just got donated a Sansui AU-517 for my website and saw the...

I just got donated a Sansui AU-517 for my website and saw the 2nd pic below in the brochure. Sansui calls this a "true DC" design and claims very wide frequency response (200,000 Hz to 0 Hz (DC) ). As a result of this insanely wide response, Sansui says this makes audio in the hearing range (20hz to 20kHz) that much better.

I'd like to hear opinions on their claims. You can find the whole brochure right here: https://zosoncsu.files.wordpress.com/2017/12/hfe_sansui_au-517_tu-517_brochure.pdf

Where could I go to learn more about purely DC amplifiers like the AU-517/717?

A 0Hz frequency response... i'd take that with a pinch of salt.! And if it can do it, for how long a period of time, and how many watts can it output the signal for. Because it clearly ain't a Crown Macro-Tech 5002VZ Amplifier.... lol

0Hz is DC. Maraschino amps are also DC coupled and MUCH higher tech.

Reply to this post

I used to have a PS Audio 200C amp that was a DC amp. Sounded great. In this context, DC just means that there are no capacitors in the signal path. One downside of the design was that if there was DC voltage applied to the inputs the amp would amplify it, risking damage to your woofers.

Which is something a designer needs to consider, in my opinion.

In a later version of the 200C PS Audio put in some sort of mechanism to nullify a DC offset, a servo maybe, but I never heard that amp.


On the back panel of the AU-517/717 you can actually separate the pre-amp from the power amp (see below).

Reply to this post

I had an AU-717 and TU-717. Sounded good. Had balls. Never was any kind of trouble. But, it was hardly anything more than above average. I ended up giving it away.

$500 on Ebay for working AU-517s. Your definition of average is always different for each person depending on budget.

And to make it even funnier, the woman I gave the system to moved overseas. So the whole thing is probably a very fancy mouse nest in the back of her parent's garage. And I'm fine with that. But really, I never much cared for the sound of it. Better than Pioneer, sure, but not on par with the early PS Audio, the Marantz 2250b and eventually a whole slew of Nakamichi stuff that I replaced it with. Some of which still resides in the living room, the glorious AV-10, and my bedroom, an amazing RE-1. But lately, I have been having a blast playing with a battery powered TPA3116D2 amp driving various vintage Altecs in my listening room. Incredibly dynamic, open and revealing, it's like listening to all of my music collection again for the first time. But, if I leave a pair of unplugged open air AKGs on my ears when doing so, I can almost hear the Sansui system...

Reply to this post

The steeper the wavefront, the higher the harmonic content frequencies. A step function has an infinite number of harmony. So a high frequency response amp gives what people call a faster amp. Attacks on notes are much more accurate. Overall sound is more detailed.

You can't hear it if it's above the audio range.

A non sinusoidal waveform such as music can be decomposed into a series of sine waves of different frequencies and a.plitudes. the stteper the wavefront, the more high frequency components are in the series. If you removes all of the higher frequency components then the waveform will not be the same shape any more. Yes, you don't hear the individual high frequency components but if they are missing the audio waveform that you are listening to isn't the same shape as the original. That's why high end amp designers make amps that have a frequency response of 200 or 250 kHz.

Of course it's different, but you still won't hear it.

roger8 sure you can design an amplifier that goes from DC to blue light (I think that was a Company slogan back in the ‘80s), but what’s the point if speakers only go from 15Hz to 30kHz?

Stewart, I believe that I already answered that question above.

Dirk, I recently had the opportunity to do an extensive (several week) audition of an amp with a frequency response up to 250 kHz and you can hear the difference. People talk about a "fast" amp, that where the fast comes from.

It has been shown that in many (most actually) cases injecting high frequency noise into the audio signal can actually increase the fidelity. I have never researched the science behind, but my source is Robert Harley "Complete Guide to High End Audio". It sounds very counter-intuitive, but from what I read it's kind of like how tube distortion is more pleasant.

roger8 yes, but the speaker will also modify the wave shape, so there’s absolutely no point in producing 100kHz tones in the amplifier, especially as they will never have existed in the source. Those ultra-bandwidth amps went the way of the dinosaurs, as they were tricky to keep stable and the bandwidth was in reality just a marketing tool.

roger8 if your comparison was sighted, then it’s moot. As the late, great Gilbert Briggs once said about bandwidth, “the wider you open the window, the more the dirt flies in”.

The speed of the amp also dictates how much phase margin is available to compensate for load reactance fighting back on the amp. It's not a static factor. Rather, if varies over frequency. There's a lot to amp design and it's typically oversimplified. The bandwidth DOES matter above 20kHz because it affects what's going on BELOW 20kHz. That's also a simplification, but a more palatable one ?

roger8 FYI: long term subjective testing is useless because our hearing adapts to any new sonic signature over time. We loose the ability to discern differences the longer we listen to the same source.

tommy76488 Please share more details.

Interesting points all. I can add that if didn't take me a long time to hear the difference. 10 minutes head to head with my $10k monoblocks was enough to recognize that there was something different.

roger8 Any test of equipment has to be carefully matched for SPL output, and performed double blind, otherwise there are too many subjective influences (both conscious and unconscious) to render an objective result.

I agree. I did match spl with a meter, I usually do. Not double blind though. I would need to find some volunteers with experienced ears to do that! I still have to say the difference was pretty apparent.

roger8 ok

That's not my photo btw, it's Jeremy.

dirk42323, yes, what the crooked ‘high end’ companies refer to as break in, is actually your brain adjusting to the new item.

stewart4831, I have thought for a while that this might be what is actually happening. At least in part.

Reply to this post

Correction - harmonics not harmony

Reply to this post

The best amp I ever heard went to 250 kHz

Reply to this post

Usually means much less phase shift within the audio range too...dabate continues about whether that is signifcant. Just watch what you put into it. And use short cables. And use that subsonic filter with records.

you cannot hear phase

I agree. Some studies don't, in very special cases.

Reply to this post

DC is good for welding but disastrous for speakers. At the AES shows Crown would demonstrate the DC 300 buy using it as an arc welder. They were originally designed to drive motors for vibration test equipment. Crown was not interested in making power amps for audio, the were a Mormon owned company that refused to let their product be used for the devil's music. They were sued and lost.

sam385, strange, I thought the founder of Crown was christian and not mormon. Actually Laurie Fincham of KEF demonstrated that excess phase shift at LF is audible, that well designed DC amps are a good idea. AFAIK most modrn amps are dc coupled.

Crown was a Mormon governed company. All employees had to belong to that religion as prayers chapel every day before work.

jamie59 DC is not the criteria, as microphones cannot capture DC nor can loudspeakers output DC but making sure that phase shift is kept to below 10 degree at 20 HZ and likewise at 20 KHz. That implies a 3 HZ to 150 KHz within 3dB.

tim788 Yet the owner was originally from the salvation army? Ok strange but I will buy it. Still think I've seen Crown amps reproduce rock and roll though!?

Crown was founded by a person who set up religious transmitters in South America indoctrinating the locals.

Yes Crown Macrotechs were huge in the rock & roll days. Horrible sounding things though

I lost 9 Jbl 2445 drivers in one night because of my SCS DC to Light amps. I never heard a thing. They were oscillating above 20K and cooked my speakers.

Not Mormon. Seventh-Day Adventist, as I recall.

bob7 as I remember it. Chapel every morning before start of work. And workers had to be of the same religion. I am talking 1970 period.

bob7 That makes sense as all the descriptions say the originator was Christian. I would have thought they would have said Mormon if he had been.

Trying to find actual religion of founder is not clear. Just some Christian bible punching group that set up Christian broadcasting in the 30s in Ecuador. I don't have the article I read in Crown company i saw 45 years ago. History has a habit of getting edited to suit the paymasters.

Ain't that the truth tim788!

I think this article indicates he was Christian. https://en.wikipedia.org/wiki/World_Radio_Missionary_Fellowship,_Inc.

jamie59 I know he was Christian. But there are many shades of Christian. Anglican Roman Catholic Jehovahs witness etc. But I remember that it was stated in 1972 article that you had to be same religion to work there because chapel was mandatory.

tim788 That I could believe!

What a bizarre deviation this thread took.? At least as of 20 years ago, Crown was still 'christian/ministry' oriented at upper management because They gave us an insanely good deal on the amps we installed in the 3500 seat auditorium at my Christian university. We were Baptist... if anyone is still trying to track down their denominational affiliation...?

It was only because someone at the beginning criticised the company's products for being Mormon, which seemed a little unfair, so I was intrigued!?

jamie59 I never criticised it for being Mormon but expressed an opinion based on what I read many years ago. The fact that at Crown in the 60s and 70s you had to be of the same religion to work there and have chapel every morning before work started.

tim788 It wasn't you!? It was someone earlier! Sorry!

tim788, I understand about the prayer stuff, a lot of that went on in those days!

Reply to this post

Those square waves look outstanding. I'd replace all the electrolytic capacitors though.

My low distortion valve amplifiers reproduce a very good square wave at 20 HZ with negligible tilt. Also HF rise time if less than 5 microSeconds. All with my transformers.

tim788 It takes a good tranasformer to work down to 20 and still work at high frequencies. The Hegeman trannys in my Citation do OK, but don't have a whole lot of power by modern standards.

gregory3377 the Citation lll had 60 watt per channel. Enough.

And i'd love to see the square waves and hear the new sound, if bybee quantum purifiers are installed... and i mean as many as possible.

mohammed7 Those square waves don't need nothing.

randy7757... Well those square waves ain't textbook.!

mohammed7 That is a good thing especially since music doesn't contain square waves. and I've seen much more expensive separate amplifiers that can't produce like that. and a bybee aint gonna make it better.

randy7757 I take it from that you haven't had the pleasure of interacting with Bybee Quantum Purifier technology. In regards to high-end audio applications.

mohammed7 Yes I'm very familiar with what they are as well as carbon nonotubes which are reasonably priced and far superior.

randy7757 Well i would disagree with carbon nanotubes being better than the current incarnation of Bybee technology.

mohammed7 That Sir is your prerogative

Indeed it is.

mohammed7 Yeah who wouldn't choose old technology for a super conductor at room temperature,lol.

randy7757 To quote yourself.... " That Sir is your prerogative". Not a stance in high-end audio i agree with though.

Well anyway, we'll agree to disagree on this point.

Mohammed Wasiq Sheikh Yeah a oxymorone always makes it right,lol.

randy7757... Looking through rose tinted glasses and having delusions or grandeur still doesn't mean you're right.!

mohammed7 Merry Christmas buddy

Happy holiday buddy.

Mohammed Wasiq Sheikh Happy Saturnalia

randy7757... Oh God, you're one of those nutjobs.!

Mohammed Wasiq Sheikh And you can't even keep on topic and you call me names? LOL You just showed everyone what you are..

mohammed7 I will be blocking you now as you've shown every sign of a troll.

... The fact that you believe i'm a troll is funny enough, but what compounds the situation further is that... you still ain't right about square waves.

randy7757 By all means do, i'm not a troll. If you did your research, you would have known this by now.

mohammed7 This is actually the first time you mentioned a square wave and all I said was something to the fact that those scope pictures are impressive. You wen't on about selling bybee filters. STFU.

randy7757 Correction: I am in agreement and do contend with you that the scope pictures are impressive although not textbook. And as for the remark of "selling bybee filters", this couldn't be further from the truth, i couldn't care less what you think or how you percieve them.

And lastly to use your oh so elegant words "STFU".!

mohammed7 Better than calling you names as you did to me so sit down and be quiet and bother someone else.

mohammed7 This is a scientific group not a elementary playground for bullies.

You do realise you're keeping this conversion going, if you have such an issue with me, i suggest you block me. Which was your earlier threat / remark.

And i am quite aware regarding what this specific group entails. And how to conduct myself to others.!

mohammed7 Yeah I know what you are.

And whom pray tell, would that be.

Mohammed Wasiq Sheikh You're a lonely man in a foreign country and the natural born citizens don't want you there. Get used to it because the citizens are paying your way, you should thank them. END OF STORY go ahead and get your last word of PRIDE in, I will not respond.

randy7757... And there we have it ladies and gents a class A racist. Some fact about me before i block you sir.

I am a born and bred Scot, living in Scotland, lots of people "actually do" want me here, as i am their native son.

And nobody pays for me.!

Happy holiday.!

Already replaced 6 of the electrolytics!

Reply to this post

the power amp circuit is a discrete power operational amplifier. It is completely DC coupled from the input connector to the speaker terminals. It is capable of being a DC amplifier. This is quite different from circuits used in early Japanese transistor amplifiers that had a single-ended power supply and (usually) an undersized coupling capacitor between the amplifier and the speaker. The Sansui 7 was one of the first receivers that I saw using this type of circuit. This would have been around 1971.

it also used discrete opamps in all of the low-level signal processing stages: RIAA amp, tone control amp.

Some of Pioneers larger receivers in the late 70s were DC coupled and sell for serious money now.

The Japanese were slow to adopt the RCA application notes for basic DC capable amplifier. The Crown DC300 was the first commercially produced true DC amp.

tim788 Are there examples of DC coupled valve amps?

Only the Luxman Direct coupled amp and the Technics Direct coupled amps. But they were not true DC input. But from first valve to output was DC coupled. So 1 Hz to 200 KHz response.

I did design many years ago but did not market a DC Valve amp. 16 PL509 output valves for 70 watt into 8 Ohm. All stages including differential input was DC. so output was 15 times the input voltage DC to 500 Khz.That was at full output. A couple of offset pots included to null output to a few millivolts. No fancy servo or solid state either.

Reply to this post


Even good things get marketed.

Reply to this post

I don't see the need for a DC amp myself. A much greater need is to protect the loudspeakers, and to make sure the amp is stable under all reasonable loads. That usually means limiting bandwidth, which can also reduce noise.

Limiting bandwidth does not affect inband noise. that is a constant.

Sadly the circuit you show for DC protection has relay contacts in series lowering the DF. I prefer crowbar methods.

Reply to this post

It does have a loudspeaker protection circuit, which would be an important thing to add to any solid state amp.

Making me feel OLD ?

Reply to this post

FET inputs on the power amps.

Reply to this post

Yeah, that's just bullshit. In fact, there are good reasons to limit the bandwidth of audio systems to the audible range. It's a shame when companies feel they have to make up nonsense in order to stand out.

Reply to this post

ethan361 I just love your candour! Merry Christmas!

Reply to this post


Reply to this post

You're ears do not ear dc or even 10 hz, but if it do amplify near dc it will do a great job from 20 hz and that is the main reason to keep it. Of course, you need the right speakers to support it

Does the same concept of headroom apply?

Reply to this post

Well any large DC voltage at the speakers can cause catastrophic damage. They are talking more about how the circuit functions, after the transformers (I assume are outputting AC to a full-wave rectifier) it basically never sees AC again. Other amps utilize AC and many capacitors, which is Sansui's whole reason for trying it this way. They say that capacitors are the biggest cause of noise in any amplifier.

I wrote an article on the amplifier when I first got it and I included the service manual and the schematic (plus the brochure above): https://hallmanlabs.com/2017/12/16/sansui-au-517-works-schematic-more-inside/

Capacitors and noise. Funny that. Boltzmann's constant does not agree. A capacitor or inductor cannot by themselves produce noise. But resistors have a theoretical noise function. So do amplifying devices.

There was a fashion in Japan for DC type amplifiers. Sansui were never really a market leader but was a successful company that got it all wrong by the 1980s.

Of course Yamaha and Sony were far bigger and also made their own semiconductors.

Back when this amp was made, they only had really bad electrolytics for DC blocking purposes. Then the FET was developed far enough to be useable as a front end device. Not only that, but someone had the bright idea to integrate two identical ones in the same package. Bingo bango, you have the basis for an amp that does not need an input DC blocking capacitor. So, they made it.

dirk42323 done with bipolar as well. Careful design makes it easy to deal with offsets and bias currents.

FET at elevated temperatures has bias current in the microAmp range.

Toshiba made the best FET for any application in the industry. Way better than American FET.

They made a dual FET in a metal package the number I cannot remember but it is in my Toshiba Databook in my office. Designed for close matching and thermal coupling to minimise drift.

I have designed DC amplifiers of many different devices. Bipolar and both types of FET and vacuum tubes Valves. All for power amplifiers can be designed to keep offsets below 5 milliVolt.

Reply to this post


Reply to this post

A wire with gain? It doesn't exist my friend!

A straight wire with gain will do a good job at amplifing the sound of one hand clapping, but the subject of this thread was DC amps not amps with magical properties.

Tony.. yes it does. My 30 year old pre amp has straight wire with gain... normal and inverted out to bridge 2 amps. 127 dB S/N Ratio and seperate headphone amp

george99 have you actually measured that 127 dB??????

The theoretical noise of a perfect 600 ohm resistor is only 126 dB relative to 1.222 volt.

george99 one other point the inverted output has one more stage of electronics.

Even 200dB technically isn't perfect.

tommy76488 what are you on about?

Guys, I was merely making a childish comment. An amp with a straight wire gain is as good as warm white wine!

tony23443 prefer Red myself.

Reply to this post

I think that by DC Amp they mean that the different stages in the amp are DC coupled. That is a plus but not a big one. The frequency range of that amp is not extradordinaire either, many audio amps can re-produce in that range. What is a problem is them saying that by going above/below hearing level that the amp sounds better, that is not so. You will only hear what your ears can hear. Often people say that the higher frequencies outside your hearing range will mix with other frequencies and creates audible frequencies that you normally would not hear. Those are intermods... a bad thing as the amp is functioning as a mixer in the higher frequencies. An amp like that should have a 20-20kHz bandpass filter right on the input to get maximum enjoyment.

Yes, and the super sonic noise will IM to produce audible artifacts. It may not be a huge effect, but it's probably measurable, depending on the amp of course.

^^^ Exactly, passing stuff we can't hear is usually a bad idea, not a good one.

Try to amplify 30Hz or below in your living room... you will spent a lot of $$

My SVS sub is more or less flat down to 18 Hz. And it has its own 1,000 watt power amp. Now, 10 Hz or lower is difficult!

1000 watts @ 18 Hz is nothing... maybe a small living room

Not small, but not huge, at 25 by 16 feet with a ceiling peak at 11 feet. But I have a lot of bass traps and other acoustic treatment, which helps solidify the bass quite a bit. My system can play at realistic rock concert levels without straining.

You say it measures flat, but at what level?

I go down to 30 Hz with a PRO PA system... all speakers are amped seperately and I use a digital xover from DBX. Filter free so as flat as a dime... at any level

Daniel Lindqvist Acoustics is mostly linear, so whatever happens in a room (frequency response, ringing) happens the same at soft and loud volumes. But no room measure perfectly flat, not even mine with a ton of bass traps and other treatment. If you can get the typical 30-40 dB span down to 10 dB, that's very good.

Reply to this post

I don't really understand the point of frequency response below or above what we can hear. Or distortion well below what we can make out. I always thought it was just engineers saying lets do it because we can.

Your brain puts what enters your ears and what your chest stomach feel as part of sound. We detect sound down below 7 HZ and also up to 40 KHz.

I am not going to go through the whole argument now.

I get being able to feel it if you have an amp and speakers that are able to do that

I believe you are right It's like a 280 kmh car, do you really will drive 280 kmh?

I mean what's 160 mph in kph lol

Eh it's only 260 kph haha

Alex.. distortion plays a major factor at high power: It can destroy your speakers

But between.1% and.05%?

Those are good numbers.. my macro-tech amps are.1 % and at times I do crank them close to max

Nobody can hear or perceive frequencies much above 20 KHz, and that's for young people. This has been researched for 50 years, and nobody has ever shown proof to the contrary. I'm glad to consider any proof! But it has to be compelling.

You can feel frequencies lower then you can hear. But mostly at high volume.

Really low low frequencies (64ft organ pipes?) can make your trousers tickle your leg hair, but we don't perceive that as musical sound. Nor ultrasonic distortion, so 2nd and higher order harmonics of 10kHz content is inaudible. It's coexisting IM difference products that are trouble - i.e. when a 40kHz distortion product interacts with one at 39kHz, producing a 1kHz "burr" sound. Best to bandpass-filter everything outside 20~20k. Both your listening level and room or medium noise affect the audibility of distortion: At background music level of say 60dB SPL even in a quiet car with 50dB ambient noise, the SNR is a meager 10dB, and unmasked distortion at 3% will only just become audible (signals 15~20dB below noise are just detectable). Played at original 85SPL in a quiet listening room of 25SPL, now an SNR of 60dB, distortion of 0.01% (-80dB) might be audible. BTW, if instantaneous peaks reach 105SPL, or 80dB SNR, signals down 15~20dB below the -93dB of a dithered CD will be more dynamic range than can be perceived. This argues that 44x16 is HD enough.

alex1 3 grotes ha ha A pilot joke. Seriously 257 kph

Yeah my dad has a 600 hp vette. It's pretty fast. I have a supercharged Lexus that's in the process of being faster

robin8748 my gosh. What???? Your trousers and feet rattling as well as you chest. For Sykes fake listen to the world with headphones. No body effect no realism.

The Deaf percussionist Evelyn Glenny deals a lot is sound.

My tests and others have confirmed people being able to detect a square wave 10 k from a sine wave. There is no second harmonic only 3rd 5th etc. Proven. You just repeat what has been told to you via Wikipedia etc.

Do some research please for yourself. Not reading stories.

The 285 Bhp of my Skyline in Japan. I can only use that power for a few seconds a day. That is zero to 40 mph at max rate then shelf cruise.

The horse power thing has become a meaningless thing in the real world as I find most Porsche drivers are wossies. Couldn't even handle a tenth of the power without loosing it. To me 100 HP per ton or zero to 60 mph in 9.5 seconds is enough on highways to loose a licence.

alex1 sadly distortion by itself does not damage speakers as music from say a piano is nothing but distortion of the fundamental that gives it its sound signature.

What you are talking about is bad clipping transistor amplifiers that cause tweeters to fail. Because of a high power of stuff above 8 KHz.

Switching between sine and square 10 KHz waves can sound different, but not because we hear ultrasonic harmonics. As robin8748 pointed out, what you're probably hearing is IM distortion that aliased back down into the audible range. This is from my Audio Expert book:

I've also seen claims proving the audibility of ultrasonic content where a 15 KHz sine wave is played, then switched to a square wave. Proponents believe that the quality change heard proves the audibility of ultrasonic frequencies. But this doesn't take into account that loudspeakers and power amplifiers can be nonlinear at those high frequencies, thereby affecting the audible spectrum. Further, most hardware generators used to create test tones output a fixed peak level. When the peak (not average) levels are the same, a square wave has 2 dB more energy at the fundamental frequency than a sine wave. So, of course, the waves could sound different.

Again, this has been researched endlessly, and whenever someone thinks they found "proof" that we can hear or perceive ultrasonics, it ALWAYS turns out to be a flaw in the testing.

The more interesting question for me is why some people are so determined to prove we can hear ultrasonic frequencies. Why is that important to people? How would that change anything?

ethan361 I know you are one of the doubters.

I really don't care what books you quote from.

My method eliminates amplifier and speaker distortion. Only ears and air can have any influence.

Simply put if detectable repeatedly then a fact.

My other test is to play 20 kHz at 112 dB. And go up in frequency from there.

As standard hearing tests are at relatively low volumes and the ears hearing mechanism is not a brick wall filter.

Sorry buddy. I really would love to put you in a room with a 500 watt ultrasonic bath cleaner at about 45 KHz the resonance of the piezoelectric drivers. And you will experience Tinitus in just a few seconds of exposure.

Challenge me at my place any day

My point wasn't to quote a book as "argument from authority," but rather to explain why switching from a square wave to a sine wave isn't a valid test. Did you not understand that a level change of 2 dB is audible all by itself?

You might also be missing that IM distortion happens readily in our ears, and the sound doesn't have to be excessively loud. After those incidents in Cuba recently, it's known that our hearing can be damaged by extremely loud ultrasonics. I'm pretty sure that was already known. But that has nothing to do with whether content above 20 KHz is necessary or even remotely useful for enjoying music.

This short test lets you identify IM distortion in you own ears at various volume levels:


Precisely. It is this a part explanation of why we hear some things. But IM distortion would happen with natural sounds. Useful or not. If it's there I require it.

Everything in the natural world of sounds has a value.

Bury your head in the sand and enjoy.

Because of people like you progress stops.

Why do I bother to make a disc cutting system demonstrably better? Why do I insist on one step pressing. Why did I bother stating that 384K/24 bit. Was needed to satisfy the hearing mechanism completely back in 1984 when CD was perfect sound forever. Two absolute words. No other interpretation of perfect or forever as in infinite.

tim788 Well now you're just being insulting, rather than sticking to the discussion at hand. That doesn't help you make your case, which seems more based on magical thinking than audio science. Look, you're welcome to believe what you want. But until you have real evidence - repeatable and preferably the result of blind testing many people over a statistically significant number of trials - you're just another person offering an opinion.

Yes Ethan. You can only quote without doing any trials yourself because you are a lazy journalist.

The pen being mightier than the sword.

What profound things have you contributed to the world of audio?

When you can say you have contributed to the world usefully then I will take back my so called insults.

OMG what an asshole!

Look, I'm not here to fight, or promote myself, or tout my accomplishments. I'm here to help people, and to educate those who want to be educated.

My Audio Expert book is used as the main text at several university and independent recording courses, and it was just nominated by the NAMM organization for a 2018 TEC award. Where's your book? I was invited to chair two Workshop presentations at recent AES shows, both of which are now videos on YouTube. What professional audio events have you hosted? My free college-level Music Theory course on YouTube has more than 1/4 million views, and thousands of Likes. Where's your music theory course? I wrote and performed my own cello concerto in public with an orchestra. Where's your concerto? I've had nearly 200 articles published in audio (and music and computer) magazines - real print magazines, not online "opinion" sites. Where are your published articles? I've designed dozens of electronic devices including equalizers, a distortion analyzer, and an entire multi-track recording console. What circuits have you designed? My Cello Rondo Music video that I performed, recorded, and video'd all by myself has nearly 2 million views on YouTube and elsewhere. What music have you performed or produced?

tim788 you're the first I've heard call the sound of a piano "distortion." Tone color (timbre), yes. But I believe the definition of "distortion" should be added coloration that is usually undesirable. Of course there are some audiophiles who swap out components seeking just the right added coloration.

That's what I said a journalist. Very good at writing books.

Don't want to take on board any controversial thinking.

Call me an asshole. Without one your dead.

That is a good compliment. YouTube wow a Wikipedia of lots of popular virals.

robin8748 well, how do you define a fundamental tone with added harmonics. So how do you define distortion. As it is not an absolute word like zero or perfect. So how do you define a perfect piano then?

So I am taking a literal meaning. Since no piano is identical there exists no perfect piano. Therefore any could be construed as having some other distortions.

Just because others don't see the world as I do does not make me wrong. If I called your mother ugly you would not be happy. If you called me ugly I would take that as a compliment.

Back to stating that a squarewave is about 2dB louder, bollox, complete and utter bollox.

A 10 K squarewave with infinite harmonics and zero phase distortion but with same peak to peak voltage. Contains the same 10 K power wise with same peak to peak. So that remains a constant. Now if you say we cannbot detect 30 K the 3rd harmonic then the higher series counts for nothing. The 3rd harmonic is less than 2 dB extra energy as read by an RMS meter. Don't forget RMS is a defined of power.

Therefore if we detect a difference it is not anything that has changed with the fundamental 10k.

Therefore to repeat my method. 3 loudspeakers and amplifiers. A 10 kHz oscillator linear and a phase locked 30 KHz oscillator and a 3 RS oscillator at the 5th harmonic also phase locked. Into the 3rd speaker. I assume all speakers can reproduce flat such as the Pioneer ribbon. So rack tone has no significant distortion but is a fundamental with less than 80 dB down other products.

This is one of the methods I did more than 40 years ago to prove what distortions are detectable.

Repeat exactly the experiment.

tim788, the difference between sine and square wave fundamental energy is very easy to prove, and I'm glad to educate you. I did this simple test in Sound Forge, and the 2 dB difference is very clear. Apology accepted.?

But that is total energy. The energy content of the 10 k remains absolutely and I repeat absolutely the same. The energy of the infinite series of harmony is the extra 2 dB but not RMS. Unless there is absolutely zero phase distortion.

Now the discussion was about only the 3rd harmonic which is not and extra 2 dB. Sorry to be pedantic. But you only can quote someone else's work not yours. Now go play somewhere else.

If you can't see that 400 Hz is louder for the square wave than for the sine wave, you're either blind or being willfully ignorant. Seeing how poorly you play with others, I'll guess the latter.

A 400 HZ is not what the fact or discussion was about. 400 HZ is close to the middle scale in music and obviously we hear the 3rd 5th 7th 9th 11th harmonics easily. And they are in a decaying scale. Well known. Moving goal posts.

Here's what I said before, which is correct but you objected to:

"When the peak (not average) levels are the same, a square wave has 2 dB more energy at the fundamental frequency than a sine wave. So, of course, the waves could sound different."

Tim, you really are an idiot. Sorry, but it's the truth. Plonk.

ethan361 so smart are you. Where is your products and contributions. Go and play your games elsewhere. So simple us your view.

LOL, I already listed some of my accomplishments. I guess you can't read either.

I'm done with you. I don't have time to deal with a jackass. You lose. But go ahead and have the last word.

tim788 "Just because others don't see the world as I do does not make me wrong."

Yes, it does if what you see is not real. If i hallucinate, does that make the pink unicorns real? What you experience in life can be both real to you and not what happens in reality. The trick here is to have enough braincells to understand the difference. I have rarely seen a moron of that caliber. You have been shown facts and you continue to argue and start calling people assholes quite fast.. I know exactly why; when we challenge your ideologies, you feel physically threatened.

I quote Ethan: "Tim, you really are an idiot. Sorry, but it's the truth. Plonk."

Truer words have been rarely spoken.

Please refrain from ridiculing each other. thanks.Please accept the fact that if you are upset about what someone says, and they did not intend to upset you, then you need to own it. You being upset about what someone else says is about you, it's not about them, unless they are trying to make you upset on purpose.

kennett7443 I do not use language like that. Stupid perhaps or poor thinking.

If someone refuses to duplicate a test but keeps quoting a source. To me that is narrow minded.

tim788 A test that tests apples and oranges need no consideration. Squarewave vs sinewave has so many issues that having difference is well expected; that is what is suppose to happen. It does not prove anything. Ultrasonic bath cleaners are not a hifi devices, they emit much more than just one band.

tim788 "The energy content of the 10 k remains absolutely and I repeat absolutely the same."

That is not correct. In a square wave, the amplitude of the fundamental is greater than that of a sine wave of the same peak value.

To illustrate, here's a quick square wave approximation comprising the 1st through 75th harmonics, overlaid graphically with its fundamental or 1st harmonic.

bob7 Thanks, maybe now Tim will believe this.

Reply to this post

Well guys/gals, I think I found a fault with this amp finally. With or without running my Kenwood C-2 Basic pre-amp the Sansui will kick into protection mode if I am giving it over 50% volume. Despite the fact there is no distortion and it's not overheating this fast. Ideas on things to check? I realized I forgot to check the transistor voltage biases.

Put a voltmeter across the speaker terminals... see if any DC is present

Also look for DC from the previous component before the pre-amp.

I'm not picking up any DC voltage across the speaker terminals of the AU-517 when it's running by itself without the C-2 pre-amp.

Reply to this post

Quite a theory...

And for those who don't already know, the db of dbx was "david blackmer", founder of both dbx, Earthworks Audio and author of the paper, "The World Beyond 20kHz".

It's complete bullshit, so no need to waste your time.

daniel7 Yes, right off, in the first paragraph, he's full of crap: "Many listeners hear a great difference when 20kHz band -limited audio signals are compared with wide band signals."

That's true only if the test aren't done properly. As it happens I just this past week completed a proper test that people can do in the comfort of their own environment. I'll announce it later today on my own Facebook page, and I'll post it here as well. Then everyone who believes they can hear the difference between 24/96 and 16/44 can learn the truth.

ethan361 Well it's obvious, the limit is in our ears. 22Hz is the usual max for humans, and that's for children. Anything "heard" above this would be distortion and definitely unwanted.

When it comes to microphones and sound sources that have high ultrasonic content; IM distortion happens if the mic can't capture the extra content. It does it at the electric conversion stage, in the transducer itself. "Jingly keys" test is common way to find out, triangle is another one that shows it quite clearly. It is a nasty kind of distortion for sure and will ruin the take, one has to implement quite drastic filtering to get rid of some of it but it is fools job; the distortion is all over the place.

We don't even hear distortion or IM aliasing up there. I used to believe that we might hear 2 KHz if 23 KHz and 25 KHz were played loudly enough for our ears to distort. But then someone who knows more than me said No, so I tested it. This was only a year or two ago. I'm 69 years old and can barely hear to 13 KHz. So I played 16 KHz and 17 KHz, which I know my big JBL speakers can reproduce. No matter how loudly I played the tones, I heard nothing.

Funny how hard you all are on the guy. I can understand why some people would perceive higher quality if the ultrasonic aspects are removed (Sansui AU-517 has an ultrasonic filter option). It is highly dependent on the speakers in use, the amp/pre-amp and your choice in music. Also you have to factor in what people "think" sounds best, even if it doesn't match up to the science of audio. Even if it isn't 100% true, it was an interesting write up. So, thanks for sharing it with us!

I'm never hard on people (unless they're insulting), but I am very hard on bullshit masquerading as science. An article that claims ultrasonic content is audible and affects what we hear is just as stupid as an article claiming that Bigfoot is real and astrologers can predict the future. It's fine as a plot for a sci-fi movie, but it's deplorable as a topic for audio science.

So what about the claims Sansui made about their 0Hz to 200,000 Hz performance due to their true DC design? They said having that kind of range only makes the audible range that much clearer since it's not stressing the system anywhere close to the maximum.

Let us say I run my DDS Signal Generator at 200 kHz into my amp, what should I expect to happen? Will I be able to hear anything at all?

keith4 "Hear?" No. See it on a 'scope. Maybe.

keith4 Not sure how much those high end signals actually does tax the system, to be honest. They don't require a lot of energy to play.

keith4 Many companies make many claims that are nonsense, hoping to fool people into thinking it's a feature that sets them apart. There is no benefit to ultrasonic, and there's no benefit for the audible range to pass frequencies above. There just isn't. In fact, there are good reasons to intentionally limit the response of amplifiers an other gear, to avoid IM distortion.

In linear power amps, -3 dB at 200 kHz isn't all that unusual. Most pro amps whose schematics I've browsed have a single pole in the output section anywhere from 159 kHz to about 250 kHz. There's usually another low-pass filter upstream in the input section.

I think response to 0 Hz (DC) is daft, through. That's like having a barometer for a microphone.

FTFY: "Even if it isn't 100% true, it was an interesting" piece of fiction.

This is audio. No room for imaginary constructs. It is like Ethan said, this is like saying a book about unicorns isn't entirely true but since it is somewhat factual, we need to pay attention to it like it is real.

What happens inside your head should stay inside your head. Before they can be really said out loud in public one should first make sure they are not spreading nonsense. We have way too much of picking and choosing your own set of facts in this world already. With audio at least it is quite simple to verify theories.

Ethan Winer I have read multiple articles on this amplifier and nobody has ever disputed their claims with their own benchmarks to back it up. I don't just believe what companies say, but in the 70's they were less likely to fluff up their specs. Also the fact that Sansui is considered by many to make some of the best vintage solid state amps.

keith4 Look, you can believe what you want, but it's total BS that an amplifier passing frequencies beyond the audio band improves audio within the audible band. I can't for the life of me understand why some people want so badly to believe this stuff. But a reputation for making good stuff is hardly scientific proof that all their advertising claims are true!

ethan361 people want to believe because they spent a helluva lot money for it. ?

Does the ability to pass ultraviolet light make for better reading classes?

keith4 I'm hard on people that make assertions (many people hear beyond 20kHz...), but provide no evidence; they follow describing the anatomy of an ear, but, really no evidence or their claim...

ethan361 Talk to Rupert Neve as to why, he actually explains it and brings devices to prove it.

Quite sure we can find Neve talking about it online, we don't need to bother him... It is quite known that Neve has that believe. Last time i checked, he still believes in that Oohashis study that if not mistaken he participated as not part of the study but did the same test. it was marred with IM distortion due to several mistakes and has been fully rebuked since.

It just proves that no one is infallible and that no matter how much experience, ego gets us all in the end.

kennett7443 What about his test where he turns on the oscillator to those frequenciesg and you can tell when its on or off not just from hearing (as I understood it) and you could tell the difference. Why not bother him, he doesn’t mind. Cognitive Bias for everyone!!!

You need a spectrum analyzer of the output from the speakers to make sure that you're not just hearing IM distortion. I'm pretty sure Neve didn't do that.

^^^ Exactly. This is from my Audio Expert book:

Years ago there was a widely publicized anecdote describing one channel in a Neve recording console that was audibly different than other channels, and the problem was traced to an oscillation at 54 KHz. I'm sure that channel sounded different, but it wasn't because Rupert Neve or Beatles engineer Geoff Emerick was hearing 54 KHz. When an audio circuit oscillates, it creates hiss and "spitty" sounds and IM distortion in the audible range. So obviously that's what Geoff heard, not the actual 54 KHz oscillation frequency. Further, no professional studio monitor speakers I'm aware of can reproduce 54 KHz anyway.

daniel7 Dont be pretty sure, lets get him in and find out definitively.

By the way, this is NOT what I’m talking about. I’m not talking about some accident that happened on a Neve board nor some cognitive bias by Neve regarding an accident. He travelled with an oscillator, amps, and transducers capable of reproduction in the frequencies he said we could hear and feel and would prove his point clearly in your face in your studio. THAT is what I’m talking about.

brian4769 There's a famous test with "oscillators, amps, and transducers" that failed due to IMD in the tweeters. This is also from my book:

There was also a study by Tsutomu Oohashi that's often cited by audiophiles as proof that we can hear or otherwise perceive ultrasonic content. The problem with this study is they used one loudspeaker to play many high-frequency components at once, so IM distortion in the tweeters created difference frequencies within the audible range. When the Oohashi experiment was repeated by Shogo Kiryu and Kaoru Ashihara using six separate speakers, none of the test subjects were able to distinguish the ultrasonic content. This is from their summary:

When the stimulus was divided into six bands of frequencies and presented through six loudspeakers in order to reduce intermodulation distortions, no subject could detect any ultrasounds. It was concluded that addition of ultrasounds might affect sound impression by means of some nonlinear interaction that might occur in the loudspeakers.

brian4769 Look, it's been well known for decades that people can't hear or otherwise perceive ultrasonics. What I don't understand is why some people are so determined to try to prove that ultrasonics are audible.

ethan361 Fantastic, so, who has THAT system In their house or studio? Is that study asserting that people CAN’T perceive or can’t hear? Huge diff.

ethan361 I’m not trying to prove anything, Rupert actually goes around and does his thing, or, he used to, I imagine he’s getting on in years. What he was trying to prove is that when you record music with a system with the capability to record and/or reproduce what was recorded including extended bandwidth audio produced by the recorded material, it can be percieved, and that music recorded without that is missing the interaction byproduct in the audio and whatever it causes the transducers to do and THAT is not only perceptible but also more representative of the “being there” important part of the reproduction. MOST people cant hear above 15khz. Does that mean we should save money, time, disk space, etc and then argue that anything above it is unimportant to most and thus imperceptible by any means? I dont think so but some do.

You miss the point. If people can hear a difference between extended highs and a filter at 20 KHz, then the test is flawed. It's that simple. This has been tested many times. When done correctly nobody can tell the difference. When done poorly, people can tell. Testing this stuff properly is not trivial! Which is why these myths persist.

Reply to this post

You need a full decade to the -3 dB point to have phase linearity.

I know in filters and control systems phase or phase linearity is a consideration, but does human hearing require phase linearity?

peter27309, Yes, In band phase response is affected by out of band frequency response!?

george6 Do not ask t hat question, #fbrother may not have disk storage for the discussion it may lead to. However some loudspeaker manufacturers do add supertweeters with a very high upper response limit to get a nicer phase response in the audible band.

peter27309 OK, I understand that manufacturers do certain things, but that doesn't prove that it is a useful or required feature for our enjoyment; it could be just a "business decision." I think we have enough resources, research and storage to examine this topic...

jamie59??? skeptical... do you have more clarification?

George Lukes It's absolutely standard knowledge encapsulated in the Bode Approximation for sketching phase and amplitude responses. For magnitude x2 and x1/2 of the cut-off freq. is sufficient. for phase however x10 and x1/10 of the cut-off freq is necessary.

jamie59 I understand electrical engineering and control systems very well, so I understand your statement: "It's absolutely standard knowledge encapsulated in the Bode Approximation for sketching phase and amplitude responses. For magnitude x2 and x1/2 of the cut-off freq. is sufficient. for phase however x10 and x1/10 of the cut-off freq is necessary." But my point/question is: Is all that necessary for the human perception and enjoyment of music? Does phase affect our perception of music? I am not aware of any human factors research that corroborates that.

The exstreme bandwidtth designs are/were a folly to fit the requirements of the marketing departments caused by Harman Kardons "square wave ads". While it does have merit to have wide bandwidth in the silicon it should be restricted to prevent pre-.and poweramplifiers from becoming AM receivers, something that tends to reduce the life expectancy of a power amplifier and occasionally endangers any treble unit affilitated with it.

george6 There is a meta study published in the AES Journal that suggest this is the case. "A Meta-Analysis of High Resolution Audio Perceptual Evaluation Document Thumbnail

Over the last decade, there has been considerable debate over the benefits of recording and rendering high resolution audio beyond standard CD quality audio. This research involved a systematic review and meta-analysis (combining the results of numerous independent studies) to assess the ability of test subjects to perceive a difference between high resolution and standard (16 bit, 44.1 or 48 kHz) audio. Eighteen published experiments for which sufficient data could be obtained were included, providing a meta-analysis that combined over 400 participants in more than 12,500 trials. Results showed a small but statistically significant ability of test subjects to discriminate high resolution content, and this effect increased dramatically when test subjects received extensive training. This result was verified by a sensitivity analysis exploring different choices for the chosen studies and different analysis approaches. Potential biases in studies, effect of test methodology, experimental design, and choice of stimuli were also investigated. The overall conclusion is that the perceived fidelity of an audio recording and playback chain can be affected by operating beyond conventional resolution.

Author: Reiss, Joshua D. Affiliation: Queen Mary University of London, London, UK JAES Volume 64 Issue 6 pp. 364-379; June 2016"

It's open access so anyone can read it.

Reply to this post

Reply to this thread

This site uses cookies and other tracking technologies to differentiate between individual computers, personalized service settings, analytical and statistical purposes, and customization of content and ad serving. This site may also contain third-party cookies. If you continue to use the site, we assume it matches the current settings, but you can change them at any time. More info here: Privacy and Cookie Policy