John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Christophe,
The enclosure and the waveguides are all molded in polyurethane structural foam. Thick walled and very well damped. I had my own molding company set up at the time that I did that. I had a very large molding press and specialized Reaction Injection Molding machine which I still have in mothballs in storage. One of my specialties is plastic mold design and materials application. I spent 20 plus years doing product development and materials application. I was molding composite parts that were up to 4'x 7' and demolding them in less than two minutes. This development is what GM used later to produce the pickup truck beds on some trucks for a few years. I also did that for some solid fuel rocket development. I did a lot of first but like John I only seemed to make my customers rich while I did all the work. I closed down my operation and went into aerospace and now I am back at the audio development that I always was doing when I had the time.
 
Blind test have nothing to do with "I like it ,or not "..They should discover really audible differences. If blind test show, that someone cannot distinguish two or more audio components by ear , it have no sense to talk about diferences in sound. That´s simple.

My point was, there are things that you cannot unleashed in a short-period blind test. How about people who cannot listen well. Do you think they should stick with LM1875 just because they cannot hear the difference with Aleph J for example? How lucky they are...

One important thing that I try to observe from a sound is musicality or fatigue or whatever that should be but related to my objective in listening to music, which is enjoyment.

A good system to me, is that sound that will keep me awake. I will stay up late, even didn't want to go to work. My wife will complaint because I stayed all day in the room listening to music. I like a system that will display who is the best singer, who are the top musicians, who are the best pianists, etc...

An average system to me, usually an over-engineered system, is those who is the same whether it is there playing music or not. I can go on reading book or do other things and forget about what is playing on my system. I only remember that the music was there when the CD reached the end of the last song.

A bad system to me, is that sound that I don't want to hear. Often time will tell if I feel disturbed with the sound. The best test is when I want to go to sleep or during my sleep where the music is playing. Many times I have to jump out of the bed to turn off the music.

And differences in sound must be acompanied with differences in signal , other way it is black magic and alchemy..

So to summarize, 2 test questions for you are (1) Are you sure that when you cannot hear difference in a one minute blind test then there will be no difference whether you hear it in one month or one year? (2) Can you explain anything about how fatigue, TIM/PIM or anything works, and how they can be measured if you think they are relevant.

Science has absolutely touched on a very large and significant part of audio. To state otherwise is incorrect.

Exaggerating or not, incorrect or not, I guess it depends on both psychological issue and hold your breath, science. To me it's just words contains information/message that will be lost if stated otherwise.

There are differences between those who think mostly with their "left brain" and those with their "right brain" (Please don't debate on the biology issue).

And you know what, musics and arts are usually associated with right brain. Those who can better appreciate tiny differences, such as emotion contents of a voice or face, body language, tone, I believe are those who have their right brain well developed.

Observe the famous "proverb": "The more you know, the less you understand". People like you may probably stand up quickly with objection. Your left brain says: "If nothing I know, nothing I can understand!" Good logic.

Some people OTOH, may find a "truth" behind above proverb. I have experienced the same thing, but not in electronics. I know almost anything related to the subject but then I feel I don't know or don't understand anything. The reason is because I found that those BIG knowledge database didn't reach the surface of the real issue. People with less knowledge OTOH may think they found the secret, may feel like a great scientist.

And in technical forum like this, you will always find some people who think and feel that they will look smart when they fight against things regarded as voodoo and non-sense. Basic psychology.
 
Last edited:
A good system to me, is that sound that will keep me awake. I will stay up late, even didn't want to go to work. My wife will complaint because I stayed all day in the room listening to music. I like a system that will display who is the best singer, who are the top musicians, who are the best pianists, etc...

An average system to me, usually an over-engineered system, is those who is the same whether it is there playing music or not. I can go on reading book or do other things and forget about what is playing on my system. I only remember that the music was there when the CD reached the end of the last song.

A bad system to me, is that sound that I don't want to hear. Often time will tell if I feel disturbed with the sound. The best test is when I want to go to sleep or during my sleep where the music is playing. Many times I have to jump out of the bed to turn off the music.
The problem, Jay, is that I've found that a setup can go from that bad system to the good, as you've described them, and back again, so very easily; the slightest problem will drag down the good to the bad in a twinkling of an eye. Which is why a lot of people make do with your average, a lot safer and more predictable; it's living on the edge for me, keeping the head of a "goodie" above water -- may the force be with me ... :)

Frank
 
The problem, Jay, is that I've found that a setup can go from that bad system to the good, as you've described them, and back again, so very easily; the slightest problem will drag down the good to the bad in a twinkling of an eye. Which is why a lot of people make do with your average, a lot safer and more predictable; it's living on the edge for me, keeping the head of a "goodie" above water -- may the force be with me ... :)

Frank

Yes, correct. It was the hardest part to learn (understanding the weakest link). My last lesson was about tube heater supply regulation. It was hard for me to accept tube "distortion" until I found out that tube heater regulation can easily become a bottleneck/weakest link.

Luckily I know more about speakers than amps. It is the "heart" of any audio system. Speakermen know more about good system than ampmen I believe.
 
It was the hardest part to learn (understanding the weakest link). My last lesson was about tube heater supply regulation. It was hard for me to accept tube "distortion" until I found out that tube heater regulation can easily become a bottleneck/weakest link.

Luckily I know more about speakers than amps. It is the "heart" of any audio system. Speakermen know more about good system than ampmen I believe.
Well called. The weakest link is where it's at, and one must make no assumptions, whatsoever, about where that can be.

I'm not proud: I call myself a systemman, I have no favourites amongst the gear; whatever's dragging the chain is part of the Evil Empire as far as I'm concerned ... ;)

Frank
 
But we are on the way and we know, at least, when two signals are identical in a decent frequency range around audio range: 10 to 100 Mhz.
Harmonic distortion, Fourrier analyses, Intermodulation analyses, phases and delays (group delay, phase modulation) etc...

Not these approaches.
In your later post, you have mentioned direct error extraction between output and proper time shifted input, two voltages at a given time moment. If somebody will manage to achieve -160dB accuracy in this way (like Ed Simon did with resistors measurements), this will be real advance in applied sound reproduction studies. Why -160dB ? Resistors are quite listenable, but their measurable artefacts are around -160dB only. Similar numbers, I guess, are associated with capacitors, contacts, wiring, etc.
 
AX tech editor
Joined 2002
Paid Member
Not these approaches.
In your later post, you have mentioned direct error extraction between output and proper time shifted input, two voltages at a given time moment. If somebody will manage to achieve -160dB accuracy in this way (like Ed Simon did with resistors measurements), this will be real advance in applied sound reproduction studies. Why -160dB ? Resistors are quite listenable, but their measurable artefacts are around -160dB only. Similar numbers, I guess, are associated with capacitors, contacts, wiring, etc.

Bill Waslo's Diffmaker software does just that. He has some test files on his website where one track classical music has a complete sousa-band buried in it at -60db and most people cannot hear it untill the main music is stripped off.

Ed Simon's article in Linear Audio Vol 0 shows resistor nolinearity varying from -60 or -80 dB to a formidable -160 dB. The facts 1) that the best resistors manage -160 dB and 2) some resistors cause audible diferences does not lead to a logical conclusion that we can hear -160 dB nonlinearity. Bill Waslo shows that you are a factor of 100,000 too optimistic ;)

jan
 
My phono input uses matched 4 x 2SK147 and 4 x 2SJ72 per channel plus cascodes - bought them from Erno 20 years ago in the golden age of jfets. What is left for designers today ? BF 852 and waiting for Linear Devices to make a P channel device or do we go over to the dark side and use bipolars. Any sign of a surface mount N or P dual low noise jfet pair or is it all cell phone and computer chips and we are doomed to trolling Hong Kong Ebay sites for fake jfets forever?

sorry to intrude - but wonder if any decent P channel SMT JFETS have become available so we have a nice complement to the BF 852/862 ? Thought it might be nice to make some buffers with them. I could not find anything searching through this thread other than the above ref.... Thanksamillion for any guidance :)
 
Well, the maximum level our ears are able to afford is 120dB spl. Where do you put those 'quite listenable' -40dB ?

I did not investigate this subject personally, but if some assumptions contradics "experiment", try to suppose what could be wrong with current axioms, and not simply discard the "experimental" result.
I could suppose, that listenability depends on circumstances. Addition of very small fraction of definite distortion COULD BE listenable (difference between two signals), while distortion signal alone - not. I do not know exactly, but we must research for getting explainations.
On the contrary, I could ask you: please, give an idea, why a resistor type used at GNFB is quite listenable, while it adds THD portion of -120 -160dB level?
I understand, that an easiest position to say, that they CAN NOT BE LISTENABLE because of 120dB of dynamic range. But they are listenable.
 
Last edited:
Ed Simon's article in Linear Audio Vol 0 shows resistor nolinearity varying from -60 or -80 dB to a formidable -160 dB.
How can we TEACH audiophiles how to land in the REAL world ?
And give them elementary notion of the orders of magnitude ?
I read that, and it is just a confirmation that resistors are perfect, while some begin to worry about the way they... sound !
 
I understand, that an easiest position to say, thay they CAN NOT BE LISTENABLE because of 120dB of dynamic range. But they are listenable.
Some resistors (carbon) adds are a little noisy (more noise than the thermal one of a *perfect resistor*). They can vary a little their values with temperature. I'm not able to ear any change of 'sound' from a metallic resistor. It is just like worrying about a temperature change of 1/100 ° of your body.
To put the things in perspective, the best loudspeakers have a distortion > -40db under the signal level.

Now, worry if you *like* to worry and convince-you that "they are listenable", we have no power against such convictions.
 
Last edited:
Bill Waslo's Diffmaker software does just that. He has some test files on his website where one track classical music has a complete sousa-band buried in it at -60db and most people cannot hear it untill the main music is stripped off.

Dear Jan, I believe it is so, but what kind of audio system did Waslo use?
With computer attached speakers, many things can happen.
To start with, one should
- use really top rank system ("acoustic oscilloscope")
- got enough experience in listenings of high quility recordings.
 
Some resistances (carbon) adds are a little noisy (more noise than the thermal one of a *perfect resistor*). They can vary a little their values with temperature. I'm not able to ear any change of 'sound' from a metallic resistor. It is just like worring about a temperature change of 1/100 ° of your body.
To put the things in perspective, the best loudspeakers have a distortion > -40db under the signal level.

We must not generalize distortions, one kind of distortion does not attract care at -40dB level, while another is listenable even at -20dB below noise floor. Therefore, in my "personal" qualitative picture of "audio world" there are "macro sound picture" and "micro sound picture".
The first is well studied, while the second exists below limits of resolution of 99% of systems that people use. Therefore, not everybody understand, what "micro sound picture" is. I do not claim, that the last is not measurable, but it requires more advanced techniques for correct measurements, than standard ones.
 
Last edited:
Bill Waslo's Diffmaker software does just that. He has some test files on his website where one track classical music has a complete sousa-band buried in it at -60db and most people cannot hear it untill the main music is stripped off.
It's a strawman test because this is just uncorrelated random noise which we are able to filter out completely. If we couldn't, we would not at all be able listen to any music faithfully without living in a fully sound-proofed room.

Repeat the test with a correlated distorted copy of the original and you'll be surprised.
 
It's a strawman test because this is just uncorrelated random noise which we are able to filter out completely. If we couldn't, we would not at all be able listen to any music faithfully without living in a fully sound-proofed room.

Repeat the test with a correlated distorted copy of the original and you'll be surprised.

+1 - I was just about to post the same, but you beat me to it Klaus :) I'm guessing this 'test' was one dreamed up by Ethan Winer? ;)
 
Status
Not open for further replies.