John Curl's Blowtorch preamplifier

[dimitri]

Well said, Jan! Thank you, buddy

 

[anatech]

Hi,
Wouldn't you agree that if the sound is accurately recreated, you will hear it as you would the original? After all, they should sound the same to the observer, no matter how "he" hears and processes the information.

My feeling is that this fact takes that variable out of the equation. The goal will always be to recreate the event (sound in this case) as close to the original as possible.

-Chris

 

[jan.didden]

Hi,
Wouldn't you agree that if the sound is accurately recreated, you will hear it as you would the original? After all, they should sound the same to the observer, no matter how "he" hears and processes the information.[snip]-Chris

No, because there's no guarantee that the processing is the same at different times and/or circumstances. That's the whole point. The same sound can be perceived differently at different times, or a changed sound can still be perceived the same. You need some independent controls or checks that need to come from outside of yourself.

Of course this has not necesarily any bearing on enjoying music....

Jan Didden

 

[anatech]

Hi Jan,
Okay, I accept your point. I also agree with the observation that people "hear" differently at different times. Background noise may have a large effect also.

I still acert that if a person has a different perception than the bulk of the population, it still holds that if "he" hears music or a noise that is indistinguishable from the original, "he' will still perceive it the same way. That's within the normal range of "his" variables.

What I am trying to get at is this. How different people hear is not an issue. If that person can not tell the difference between real or reproduced sounds, it is accurate. If you make the same sound as the original, many people can hear it differently, but they will all hear it the same way they normally would hear the original.

Trying to figure out how a person hears and processes a sound does not have any effect on the goal of reproducing that sound or sounds. This is a path that will lead nowhere.

Besides, the plane takes off. 😀 😉

-Chris

 

[PMA]

Just one of my little interjectional notes:

Understanding how to make the circuit behave is well and good, but understanding how the ear hears is ~easily~ half the battle.

When you focus on that, and 'grok' it, then such understanding will begin to show how a single ended tube amp with it's rather high distortion figures can be found to sound excellent to the vast majority of listeners.

I am not sure if this (SE tube excellent sound) is generally accepted. To me, the answer is NO. I assume myself as an experienced listener 🙂D ), of both live and reproduced classical music. I very often listen to orchestral classical music, and visit concerts. This kind of music reproduced by tube SE results in a "sound nonsense", without resolution, and a lot of sameness. This is the field where nonlinear distortion of tube SE definitely takes place and defines the sound result.

 

[john curl]

What I have found out, at least, over many decades, is that there is more to what we can really hear than what we can measure. I have also found that academic education and 'facts' tend to change over time, and one can be trapped by a 'fact' first learned while being educated, that has to be unlearned with experience.

 

[PMA]

What I have found out, at least, over many decades, is that there is more to what we can really hear than what we can measure. I have also found that academic education and 'facts' tend to change over time, and one can be trapped by a 'fact' first learned while being educated, that has to be unlearned with experience.

I agree, completely.

 

[john curl]

I also want to point out that it is important to give listening the dominant factor in a design, rather than 'good engineering principles'. Of course, most good engineering is useful and effective, BUT this is where we can be trapped by thinking that IF we made something that measures well, it must be good, and our listening must be flawed.
Of course, many professors would like to convince us that we really can't hear any difference, and therefore it doesn't matter what we do as far as audio design is concerned, and they devise tests to 'prove' that we can't hear the difference between components in their test. This, then, gives you MP-3, Dolby digital, etc, and even CD reproduction.
I have always said: "Trust your ears" That is what has made me successful in this business, and also listen to others and their opinion. Sometimes a person is so wrapped up in their own design decisions, that they are not objective about what they are designing. It is better to have others evaluate your accomplishments, rather than you, yourself. However, you might still be pretty darn objective about others efforts, especially in areas that you are not designing, such as loudspeakers, D-A, phono stages, or whatever you have not really applied yourself to.

 

[h_a]

Understanding how to make the circuit behave is well and good, but understanding how the ear hears is ~easily~ half the battle. When you focus on that, and 'grok' it, then such understanding will begin to show how a single ended tube amp with it's rather high distortion figures can be found to sound excellent to the vast majority of listeners.

This reasoning falls now and then out of the sky, but I'm still waiting for _solid_ facts behind that statement ("understanding how the ear hears").

Give me some references please, I cannot find anything related in my recording engineer lecture notes.

Have fun, Hannes

 

[Juergen Knoop]

200kHz.
100kHz (-3dB) RC yes/no is audible, in a very good system.

would this apply to each amplifier (pre/power) alone or would you suggest 200kHz (-3dB) as system approach?
regards

 

[lumanauw]

I use 250khz (example : 2k2+270pf) low pass filter

 

[AndrewT]

I like 160kHz to 230kHz (rc=1uS to 0.68uS) as the low pass filter on my power amps.
I try to select/set the LP filter on pre-amps at least an octave above this (0.5us to 0.33uS).

 

[jan.didden]

Hi Jan,
Okay, I accept your point. I also agree with the observation that people "hear" differently at different times. Background noise may have a large effect also.

I still acert that if a person has a different perception than the bulk of the population, it still holds that if "he" hears music or a noise that is indistinguishable from the original, "he' will still perceive it the same way. That's within the normal range of "his" variables.

What I am trying to get at is this. How different people hear is not an issue. If that person can not tell the difference between real or reproduced sounds, it is accurate. If you make the same sound as the original, many people can hear it differently, but they will all hear it the same way they normally would hear the original.

Trying to figure out how a person hears and processes a sound does not have any effect on the goal of reproducing that sound or sounds. This is a path that will lead nowhere.

Besides, the plane takes off. 😀 😉

-Chris

Hmmm. I don't think we disagree Chris, but my angle is more that all this perception stuff is very individually. Sure, the most glaring differences will probably be heard by all but the pathologically deaf, but in hi-end audio we are concerned with subtle, fleeting differences. And when it is that personal, you can't really port one persons' experience and perception to another. This is the bottom line of 'subjective', isn't it: valid for ME only.

If you want to design equipment that sounds 'real' for everybody, you have, IMHO, two choices:

- convince everybody that it does sound 'real': that's the marketing angle, or
- design your equipment as transparent and neutral as you can so what comes out is what goes in and therefore is 'real'.

I think though that the first choice is the easiest 😉

Jan Didden

 

[Bob Cordell]

Most likely. According to my definition, a linear and time invariant circuit is called a minimum phase circuit if, and only if, the circuit transfer function and its inverse are both causal and stable.

Causality and stability imply that all poles of the transfer function H(s) must be strictly inside the left half s plane. Adding the condition regarding 1/H(s) leads to the requirement that both the zeros and the poles of a minimum phase circuit must be strictly inside the left half s plane.

Of course, for a passive circuit (and neglecting any effects like dielectric absorbtion) the minimum phase condition can be met for certain topologies. However, according to the definition, a non-linear or time variant circuit can't be of minimum phase and to the extend we are concerned here, audio circuits are both non linear and time variant. As a result, the relationship of magnitude response to phase response using the Hilbert transform 1/PI*int(x(tau)/(t-tau))d(tau) where int is from +/-infinity does not hold.

If you are thinking of Spice simulations, the AC analysis is always running on top of a linearized model. An AC analysis can lead to a minimum phase circuit, however this is not necessary valid when nonlinearities/large signal models are considered.

Yes, I think we are largely in agreement. However, the role played in the error of approximation by nonlinearity is a matter of degree. In highly linear circuits like we use here, the minmum phase approximation is likely to be very good, and minimally compromized by nonlinearity. Hence, under those conditions the phase can be fairly accurately predicted from the amplitude response.

I agree completely that in the presence of SUBSTANTIAL NONLINEARITY, all bets are off.

Cheers,
Bob

 

[anatech]

Hi Jan,
I feel we agree also. Probably almost completely. My next statement takes into account personal differences.

My belief is that if we can reproduce a sound perfectly, then it would sound no different than the original to everyone. All should agree, personal prejudices aside. At this time, we can get very close to perfect, but it isn't perfect. People vary in what compromises they can accept or not. Social pressure is the scattering force that affects the data for any studies, so don't expect a large correlation if a study is done. So, my point is that some people are more sensitive to certain short comings than other people. That determines what technology they like, or the voicing.

Is it possible too that the listening room has an overriding effect? I think so. We don't correct the problem, we play with components and wire to attempt a solution.

Hi John,

I also want to point out that it is important to give listening the dominant factor in a design, rather than 'good engineering principles'.

We might agree here, I'm not sure. Design by ear is not successful. Often you end up with unreliable equipment or higher distortions than necessary. There are a few "designers" that design by ear. This by their own admission. My experience has been that as you correct their engineering errors, the distortion drops as a side benefit. That and the customer is more assured of music instead of smoke when he powers up. What I think is that you apply proper engineering almost on a subconscious level as you design. You are focusing on the sound quality as the engineering is automatic for you. From experience, I can not agree with your statement as it stands. For you, it isn't even true because you engineer on automatic, without thinking of it.

My take on things is that both listening and measurements (engineering) guide each other. By considering both, you stay on track. By ignoring one or the other, you will miss the mark (or burn up).

-Chris

 

[john curl]

It is NOT that we should just 'design by ear'. I certainly don't. In fact, I rarely listen to my designs as they are developed. However, I have learned, over decades, what seems to be important in design, so that I avoid the pitfalls of 'good engineering' such as extremely high negative feedback, just to get another decimal point lower on the THD, or rationalizing that the ear cannot hear anything over 20KHz, or that we are measuring the right things, that the ear is REALLY sensitive to.
It amazes me when people try to tell me about the beauty and perfection of op amp based feedback. I, too, believed in it more than 40 years ago. I have been using IC op amps since 1966. What I learned in 1971, is that IC's did not sound as good as tubes, even if they MEASURED better than typical tube circuits that we had available. I learned the hard way, I failed to make a good sounding mixing console, even with the best IC's available. In order to 'grow' I had to go back to school, learn more esoteric circuit theory, much like we discuss here. This is where I learned that negative feedback could give your circuit more higher order distortion than what it started with, and I had to rethink WHY tubes sounded better than solid state, in general, and certainly better than IC's, even if tubes measured sort of lousy, and were audible in some ways, yet still more 'musical' and uplifting.
One factor that I found in tubes was reduced amounts of negative feedback. Why was this? Well, it was mostly because the tube circuits oscillated because of cap and transformer coupling if you put more than 20dB or so feedback into the circuit. Direct coupled tube circuits, while necessary for some instrumention, were normally not used for audio circuits. It was seemingly pointless. Also, I obviously saw that solid state circuits generated much more higher order distortion than tubes. There were several reasons for this, but firstly, tubes are naturally more linear than bipolar transistors, and even fets. The same circuit normally OK with tubes, completely failed with solid state, because the distortion levels were MUCH GREATER, unless you could find a way to add a lot of extra negative feedback to get the levels reduced to something reasonable. This is what early IC's did, by removing coupling caps and tranformers. Then, 60 dB of feedback was practical, and often you could make circuit designs with IC's or discrete solid state devices that measured better in many standard measurements, such as SMPTE IM distortion that was the standard in the '50-'70's, than the tube circuits they replaced. After all, if you could make a design that measured less than .005% IM, how could it be audible?
Then, over the years, we learned about TIM, IIM, PIM, and Hirata distortion. We also found that many passive components optimized for solid state operation, like capacitors, generated lots of distortion as well, just outside the measurement bandwith that worked so well with tubes. We have had to eliminate these added distortions, before solid state could compete with tubes, and we are still in competition today with the best tube designs, as they have also improved.
If any of these distortions are disregarded, as being unimportant, a circuit designer will find that their circuit 'might' measure well with their standard test equipment, yet sound compromised or even unlistenable by many, without any seeming reason. That is when the designer usually condemns the audio critics as crooked or crazy.
😀

 

[jam]

Mr.Curl,

That was a long post.........................you seem to have the crowd stumped!

Well, how about them Redskins!

Jam

 

[syn08]

I have been using IC op amps since 1966. What I learned in 1971, is that IC's did not sound as good as tubes, even if they MEASURED better than typical tube circuits that we had available. I learned the hard way, I failed to make a good sounding mixing console, even with the best IC's available.

John,

A friendly reminder: this is the year of 2008 and perhaps it's time to reconsider your 40+ years old experience and results.

 

[mlloyd1]

john has mentioned several times he is "playing around" with the AD797

mlloyd1

John,

A friendly reminder: this is the year of 2008 and perhaps it's time to reconsider your 40+ years old experience and results.

 

[john curl]

I have 100's of different IC op amps in my lab. I am designing with them at this time. They still have their problems, just like the old days. Too much feedback, a nonlinear transfer function, etc.