I'm going to present in my ignorant best and suggest that any circuit that requires the capacitors to "burn in" is badly designed for that particular capacitor.
Never having tried to compare before and after "burning in" I must be talking through a hole in my hat.
Never having tried to compare before and after "burning in" I must be talking through a hole in my hat.
Well, Andrew, I have to say that there is not much leeway in a chip amp circuit, like the one laid out in the LM3886 datasheet, for design alterations concerning the caps. The cap kind of has to be where it is, but of course many different values have been tried. I have built 5 of these things and every one of them sounded different after a week of playing. Maybe its the caps, maybe its the input resistor, maybe the chip itself needs some time to do whatever it does. But I think its the caps because if you change caps after a week it will again sound very different and mellow out after another week. If you build one you will hear it. I guarantee that.
Uriah
Uriah
madisonears said:
These are ALL audiophile sites, with subjective results. No actual testing at all. No THD, or distortion tests. Just "better""brighter" words that not only mean different things to different people, but dont actually mean anything all when talking about sound apart from vague subjective references.
You are certainly right. Perhaps you would be willing to describe the sound you hear to us in numbers. If you could give that special sound a THD number please do so. By the way, make sure you tell us which harmonic distorted so we know whether that number sounded good or bad, or was it all the same to you?
Uriah
Uriah
What im saying is this, "the sounds is grainy, and cold, but accurate"
That means nothing at all. What does grainy MEAN exactly?
That means nothing at all. What does grainy MEAN exactly?
Look, there is a feeling you get when you listen to music. That feeling can be like I mentioned a page or so back.. when I felt the music made the air feel thick and claustrophobic. Exactly how shall I measure that? We talk like this and we understand the words as they are communicated. Most communities have words common to them that have a definition the community understands. We get it when someone says the highs were 'airy'. We get it. Concept successfully communicated. And the measurement can be taken and be exactly the same as the cap used before that sounded muddy. So, its subjective. Okay, where in audio did someone say the sound wouldnt be subjective? Thats frankly a given. He likes tubes, I like solid state. My measurements show that my sound is more a real relfection of the CD content than his. Who cares? He likes it better that way and he can say it sounds smooth and warm and I can know what he's talking about and say, yeah, but I dont like it smooth and warm I like it crisp and analytical which is complete heresy to him. So the numbers mean nothing in that situation but the words convey the meaning very well.
Uriah
Uriah
I guess i always thought that audio amplication and playback was aimed towards exact reproduction. What goes in, comes out as close as possible to what was recorded.
The only way to measure that is with distortion figures, ya know?
Like if you compared the output signal with the input signal, if there is low distortion and THD, that means that the output signal is faithfully following the input signal.
For me i guess, i aim for reproduction. Not something that sounds "nice" im after "Exactly as recorded"
I think its because im a computer tech, and i work in logical stuff a lot, and with electronics as well. And im also a gamer, and with positional audio, you want as close an accurate reproduction as possible.
And that IS measurable with numbers.
Thats why people use op amps with super low distortion figures right?
The only way to measure that is with distortion figures, ya know?
Like if you compared the output signal with the input signal, if there is low distortion and THD, that means that the output signal is faithfully following the input signal.
For me i guess, i aim for reproduction. Not something that sounds "nice" im after "Exactly as recorded"
I think its because im a computer tech, and i work in logical stuff a lot, and with electronics as well. And im also a gamer, and with positional audio, you want as close an accurate reproduction as possible.
And that IS measurable with numbers.
Thats why people use op amps with super low distortion figures right?
Gentlemen,
This isn't the thread to discuss objective vs. subjective listening impressions, especially when neither party has a chance in hades of convincing the other differently than they already believe.
Agree to disagree already.
I have been waiting to build my amps to see what tweaks are going to be worthy. My intent is to build as supplied and allow it to run awhile ( call that period what you will ) before I decide if it needs any changing. I for one really appreciate Peter's volunteering to sort the various replys, and try to condense it down for us. Ultimately we all make our own decisions as to what sounds good to each of us. If what sounds good to me, sounds good to you too, well good on ya. If the reverse is true, just smile and change the subject.
I would hate to see all the effort put forth by everyone involved devolve into a "P***ing match" about burn in.
End of rant, thanks for your patience.
Keep tweaking!
John
This isn't the thread to discuss objective vs. subjective listening impressions, especially when neither party has a chance in hades of convincing the other differently than they already believe.
Agree to disagree already.
I have been waiting to build my amps to see what tweaks are going to be worthy. My intent is to build as supplied and allow it to run awhile ( call that period what you will ) before I decide if it needs any changing. I for one really appreciate Peter's volunteering to sort the various replys, and try to condense it down for us. Ultimately we all make our own decisions as to what sounds good to each of us. If what sounds good to me, sounds good to you too, well good on ya. If the reverse is true, just smile and change the subject.
I would hate to see all the effort put forth by everyone involved devolve into a "P***ing match" about burn in.
End of rant, thanks for your patience.
Keep tweaking!
John
That's all correct Rainwulf, however... people may like certain "inaccuracies" or distortions. For example, tube amps, or certain capacitor choices. Everyone has certain sound qualities they like and don't like, and exact reproduction is just one goal out of many when building/buying equipment.
In my opinion (take it with a grain of salt or completely ignore it if you like), aiming for perfect reproduction is the definition of an audiophile. It is also close to impossible. I gave up on that goal long ago, as soon as I realized that. Fortunately it was before I started spending any money on trying to achieve it 🙂 My goal now is clarity and low distortion, but at easily (read: inexpensively) achievable levels 🙂
In my opinion (take it with a grain of salt or completely ignore it if you like), aiming for perfect reproduction is the definition of an audiophile. It is also close to impossible. I gave up on that goal long ago, as soon as I realized that. Fortunately it was before I started spending any money on trying to achieve it 🙂 My goal now is clarity and low distortion, but at easily (read: inexpensively) achievable levels 🙂
Mmmm
Thank you all.
Given me a bit to think about, and also made me decide to get better speakers!
I have a 5 channel GC, yet very average 4 inch drivers in my surround speakers.
With all my fascination with accuracy, that cant be good!
Thanks again.
Thank you all.
Given me a bit to think about, and also made me decide to get better speakers!
I have a 5 channel GC, yet very average 4 inch drivers in my surround speakers.
With all my fascination with accuracy, that cant be good!
Thanks again.
Rainwulf:
This might be far off-base, but if you have access to a library that has back issues of The New Republic, look up an article in the 30 December 1985 issue written by Edward Rothstein called "The Quest for Perfect Sound." At the time, Mr. Rothstein was the much-heralded music critic for The New York Times. The article is a lengthy and really fascinating assessment of new technologies (the digital revolution had only just begun), the history of recorded sound, the lingua franca of music and the art of high end audio. Most importantly for this thread, the article speaks eloquently about the disparity between technical specifications and the quality and resolution of recorded music.
I know this is really obscure, but the article is brilliant and well worth your time. It offers a stirring perspective on the questions being debated here.
Regards to all,
Scott
This might be far off-base, but if you have access to a library that has back issues of The New Republic, look up an article in the 30 December 1985 issue written by Edward Rothstein called "The Quest for Perfect Sound." At the time, Mr. Rothstein was the much-heralded music critic for The New York Times. The article is a lengthy and really fascinating assessment of new technologies (the digital revolution had only just begun), the history of recorded sound, the lingua franca of music and the art of high end audio. Most importantly for this thread, the article speaks eloquently about the disparity between technical specifications and the quality and resolution of recorded music.
I know this is really obscure, but the article is brilliant and well worth your time. It offers a stirring perspective on the questions being debated here.
Regards to all,
Scott
I will see if i can locate it.
I honestly always thought it was about the specs, obtain the lowest noise floor, the highest SN ratio, the lowest THD, and the flattest response, and thats what gives you best audio.
I honestly always thought it was about the specs, obtain the lowest noise floor, the highest SN ratio, the lowest THD, and the flattest response, and thats what gives you best audio.
AndrewT said:
Caps do change their sound in the first hours, caps forming for elcos is a fact... 😉
Quite all caps (except stacked films) are wound; when a magnetic field is applied to plates they moves so wouldn't be so strange to expect that they settles in a different position than the original one... do that for some hours and plates find a stable position...
It's only a speculation but...😀
Rainwulf said:
Sure, but frequency distorsion isn't the only measure...
Frequency response, Phase distortion, Phase Delay and Transient Response are also important.
Also THD is often measured at a single frequency (like 1KHz) and not over the entire spectrum.
Often amplifiers with the same measures of THD sounds differently...
And every single component has different measures of such metrics depending on construction, materials, parasitic inductance and impedance (that also varies with frequency...).
Real world is more complex than the model that represents it, by definition. 😉
There are a lot more figures that go into it than distortion. One would be phase angle. That will make things sound VERY different but might not be distorted. Distortion can be a desirable attribute for an amp. It depends on how much and in which harmonic. If I am right the F5 was built so that most of the distortion would be second harmonic. Its a very analytical amp sooooo, dont go out and buy your amp because it says .0001% distortion. It might still sound like junk. The soundstage might be compressed to the point that it makes you squirm every time you listen to it.
So, are you saying that if the overall distortion is so low that it is not humanly perceptible than the amp is of the same sonic quality of any other amp with imperceptible distortion? Yes..? No...?
That statement would ignore bandwidth and noise. It would ignore intermodulation and crossover distortion. It would ignore the fact that factory distortion measurements are usually made at only 1kHz and that if your amp isnt fast enough to deal handily with high frequency high amplitude transients it'll momentarily change the NFB loop and for a second it will lose control of the output signal. So suppose you have a bunch of high frequency transients, um, like a cymbals and snares. They might sound like junk, but the distortion numbers would be nice. How about if your caps are slow and your amp gets these high frequency transients? Well, at least it measured good at 1kHz. In this amp the opamp was specifically chosen to attack this problem because it has a fantastic slew rate. So we can avoid slew induced distortion, but thats not Harmonic Distortion.
So, we get back to caps. A cap has to be fast if it has anything to do with signal. Even if all it does is facilitate the functioning of the other components in the signal. We care about C13 a lot because it is the cap that keeps DC out of the system but it is also directly in series with the signal. So we need to keep capacitor distortion down. This C13 also effects which frequencies get through and being in the signal path has the opportunity to color the signal before its amplified which means that this is probably the most important cap in the circuit. Its speed, dielectric, capacitance and resistance all have measurable changes to an audio signal, but which one will sound better? Electrolytics just dont sound as good as films... But, in some aspects they perform better electrically. The best way to avoid most capacitor distortion in the coupling cap is to shoot for a 10Hz rolloff that will throw the subsonic distortion down into the few Hz region and help to avoid coloration. If thats what you are shooting for. Some people are going to like it though.
When we talk about soundstage, presence, depth we are talking about the phase response. Its measureable and saying it has a wide soundstage is nothing weird and audiophylic (making up words again, but people still getting what I am saying) and again its not THD. So there is phase distortion. The source produces a signal and the amp either keeps the signal in absolute phase or it 'twists' it a certain degrees which will force certain frequencies to cross at certain points in the 3D space around your listening area and at that point the frequencies will cancel each other. Same reason you cant hear a BlackHawk helicopter until that final second. And then of course if your amp is perfectly in phase with the source, which is doubtful because of preamps and tone controls which twist phase, your speaker is still going to mess with the phase linearity before the music reaches your ear, but your amp might measure well.
Anyway, most of that is out of the Audiophile's Project Sourcebook by Randy G Slone. He is the expert not me, but tell me that you still think THD is the end all be all. You actually want even order distortion and any odd order distortion will sound like junk, but THD means TOTAL Harmonic Distortion and doesnt specify which order or orderS are distorted. A cap does way more than distort or not distort and it might be measureable but try 5 different caps, measure everything you can and tell me which measurement makes it sound more like butterscotch than peppermint or whatever it ends up sounding like to you because it sure wont sound like a number and you will have to use some subjective words to describe it and then all your measurements will be poopooed by some guy who says your words mean nothing without numbers and then when you produce numbers he will say 'yeah, so, whats it sound like' and this discussion would have to move to a new thread so we can battle forever while someone else listens to great music on a wonderful amp with caps that sound like honey.
Uriah
So, are you saying that if the overall distortion is so low that it is not humanly perceptible than the amp is of the same sonic quality of any other amp with imperceptible distortion? Yes..? No...?
That statement would ignore bandwidth and noise. It would ignore intermodulation and crossover distortion. It would ignore the fact that factory distortion measurements are usually made at only 1kHz and that if your amp isnt fast enough to deal handily with high frequency high amplitude transients it'll momentarily change the NFB loop and for a second it will lose control of the output signal. So suppose you have a bunch of high frequency transients, um, like a cymbals and snares. They might sound like junk, but the distortion numbers would be nice. How about if your caps are slow and your amp gets these high frequency transients? Well, at least it measured good at 1kHz. In this amp the opamp was specifically chosen to attack this problem because it has a fantastic slew rate. So we can avoid slew induced distortion, but thats not Harmonic Distortion.
So, we get back to caps. A cap has to be fast if it has anything to do with signal. Even if all it does is facilitate the functioning of the other components in the signal. We care about C13 a lot because it is the cap that keeps DC out of the system but it is also directly in series with the signal. So we need to keep capacitor distortion down. This C13 also effects which frequencies get through and being in the signal path has the opportunity to color the signal before its amplified which means that this is probably the most important cap in the circuit. Its speed, dielectric, capacitance and resistance all have measurable changes to an audio signal, but which one will sound better? Electrolytics just dont sound as good as films... But, in some aspects they perform better electrically. The best way to avoid most capacitor distortion in the coupling cap is to shoot for a 10Hz rolloff that will throw the subsonic distortion down into the few Hz region and help to avoid coloration. If thats what you are shooting for. Some people are going to like it though.
When we talk about soundstage, presence, depth we are talking about the phase response. Its measureable and saying it has a wide soundstage is nothing weird and audiophylic (making up words again, but people still getting what I am saying) and again its not THD. So there is phase distortion. The source produces a signal and the amp either keeps the signal in absolute phase or it 'twists' it a certain degrees which will force certain frequencies to cross at certain points in the 3D space around your listening area and at that point the frequencies will cancel each other. Same reason you cant hear a BlackHawk helicopter until that final second. And then of course if your amp is perfectly in phase with the source, which is doubtful because of preamps and tone controls which twist phase, your speaker is still going to mess with the phase linearity before the music reaches your ear, but your amp might measure well.
Anyway, most of that is out of the Audiophile's Project Sourcebook by Randy G Slone. He is the expert not me, but tell me that you still think THD is the end all be all. You actually want even order distortion and any odd order distortion will sound like junk, but THD means TOTAL Harmonic Distortion and doesnt specify which order or orderS are distorted. A cap does way more than distort or not distort and it might be measureable but try 5 different caps, measure everything you can and tell me which measurement makes it sound more like butterscotch than peppermint or whatever it ends up sounding like to you because it sure wont sound like a number and you will have to use some subjective words to describe it and then all your measurements will be poopooed by some guy who says your words mean nothing without numbers and then when you produce numbers he will say 'yeah, so, whats it sound like' and this discussion would have to move to a new thread so we can battle forever while someone else listens to great music on a wonderful amp with caps that sound like honey.
Uriah
I go away for an afternoon and look at all the BS being thrown about!
To all the cynics and technicians: please go study your graphs and charts and numbers and leave us alone. If we hear something we hear it, and your arguments mean nothing. If you can't or don't want to hear it, you have nothing to contribute. Let us wallow in our meaningless comparative tests and subjective audiophile babble. It's a secret code, and you'll never decipher it, but basically it means that when a trumpet sounds like a trumpet, we don't give a hoot about the third harmonic distortion level. If you can distinctly hear every tone of a large chorus's singing, and the highest female voices don't have that hazy, gritty overlay to them, then your stereo is doing its job. Try to measure that!
Andrew T, I've seen hundreds of your posts here and have the utmost respect for your technical abilities. However, if you are stating that any circuit with new components in it that changes over the course of the first few hours of playing must be defective, then, not only are you talking through a hole in your hat, you must be talking through another hole located much lower on your body. Please stick to the technical stuff and leave things like music to those of us who actually listen.
Rainwulf, I don't recognize your moniker, but it's nice of you to chime in here with your "numbers are god" claptrap. Do you purchase or otherwise evaluate equipment by looking at numbers and listening to test tones, or do you ever actually listen to music through it? It seems that you need to gain some experience before laying down the laws.
We are discussing the construction of a particular design of amplifier and the means by which we can extract the best, or rather, most musical performance from that design. We are not at all interested in graphs or numbers. The designer already did an excellent job of that. We build something, we listen. We change something, we listen some more. Call that what you will, even foolish perhaps, but if our goal is listening to music and not the ultimately accurate reproduction of an input signal, then that's a system that works. Whatever sounds more like music coming out of the speakers is what's best, whether it's full of distortion or noise or other anomalies. There is as much art to designing and building audio equipment as there is science. The science part is already done here.
Peace,
Tom E
To all the cynics and technicians: please go study your graphs and charts and numbers and leave us alone. If we hear something we hear it, and your arguments mean nothing. If you can't or don't want to hear it, you have nothing to contribute. Let us wallow in our meaningless comparative tests and subjective audiophile babble. It's a secret code, and you'll never decipher it, but basically it means that when a trumpet sounds like a trumpet, we don't give a hoot about the third harmonic distortion level. If you can distinctly hear every tone of a large chorus's singing, and the highest female voices don't have that hazy, gritty overlay to them, then your stereo is doing its job. Try to measure that!
Andrew T, I've seen hundreds of your posts here and have the utmost respect for your technical abilities. However, if you are stating that any circuit with new components in it that changes over the course of the first few hours of playing must be defective, then, not only are you talking through a hole in your hat, you must be talking through another hole located much lower on your body. Please stick to the technical stuff and leave things like music to those of us who actually listen.
Rainwulf, I don't recognize your moniker, but it's nice of you to chime in here with your "numbers are god" claptrap. Do you purchase or otherwise evaluate equipment by looking at numbers and listening to test tones, or do you ever actually listen to music through it? It seems that you need to gain some experience before laying down the laws.
We are discussing the construction of a particular design of amplifier and the means by which we can extract the best, or rather, most musical performance from that design. We are not at all interested in graphs or numbers. The designer already did an excellent job of that. We build something, we listen. We change something, we listen some more. Call that what you will, even foolish perhaps, but if our goal is listening to music and not the ultimately accurate reproduction of an input signal, then that's a system that works. Whatever sounds more like music coming out of the speakers is what's best, whether it's full of distortion or noise or other anomalies. There is as much art to designing and building audio equipment as there is science. The science part is already done here.
Peace,
Tom E
everybody CHILL
Let's get back to the thread title...
The subject was: suggestions for things to try in terms of different components.
BTW, sadly, nobody has taken me up on my offer to collect suggested exchanges. (The stream of haggling back and forth with "this might be better" "no, it's not" doesn't yield a suggested list of possible changes...)
Here is what I have in mind. If I (or someone else) wants to learn about this they (I) will naturally ask: if I want to see the biggest difference, what should I change first and with what? Let's start with something simple like that.
So... To all the people with actual tweaking experience under their belt for the revC, for the rest of us, suggest your one favorite ("this makes the most and most noticeable difference") change. I will collect.
(Or correct me in a PM if you think I am approaching this all wrong; I for one want to experiment, but I am not prepared to spend hundreds of $s on many many caps with each having diametrically opposed proponents; I don't have the patience for that many well controlled experiments...)
Anyway, I am ready to learn.
Peter
Let's get back to the thread title...
The subject was: suggestions for things to try in terms of different components.
BTW, sadly, nobody has taken me up on my offer to collect suggested exchanges. (The stream of haggling back and forth with "this might be better" "no, it's not" doesn't yield a suggested list of possible changes...)
Here is what I have in mind. If I (or someone else) wants to learn about this they (I) will naturally ask: if I want to see the biggest difference, what should I change first and with what? Let's start with something simple like that.
So... To all the people with actual tweaking experience under their belt for the revC, for the rest of us, suggest your one favorite ("this makes the most and most noticeable difference") change. I will collect.
(Or correct me in a PM if you think I am approaching this all wrong; I for one want to experiment, but I am not prepared to spend hundreds of $s on many many caps with each having diametrically opposed proponents; I don't have the patience for that many well controlled experiments...)
Anyway, I am ready to learn.
Peter
Peter,
Thanks for refocusing. I will gladly contribute to a structured database when my experiments prove more conclusive. Right now, I'm still evaluating changes that, according to the experts, can't possibly be taking place. These little Frankensteins are not quite ready to see the light of day, so back to the laboratory. (Insert maniacal laughter here.)
I'll only reiterate what I've already posted: C13 (was it designated 13 for a reason? hmmm...) must be changed if you hope to derive anything better than merely okay performance from this marvelous design. There is so much potential here, and so much payback of even a modest investment.
Peace,
Tom E
Thanks for refocusing. I will gladly contribute to a structured database when my experiments prove more conclusive. Right now, I'm still evaluating changes that, according to the experts, can't possibly be taking place. These little Frankensteins are not quite ready to see the light of day, so back to the laboratory. (Insert maniacal laughter here.)
I'll only reiterate what I've already posted: C13 (was it designated 13 for a reason? hmmm...) must be changed if you hope to derive anything better than merely okay performance from this marvelous design. There is so much potential here, and so much payback of even a modest investment.
Peace,
Tom E