I said a mild time delay when negative feedback . I should have said miniscule / non extant . Bad habits hey .
Not only interesting, it resembles the situation roughly 60 years ago, as afair, people did favour music bandlimited to ~8 Khz compared to full bandwidth material (~15Khz) during reproduction.
Olson did some quite impressive tests to investigate if listeners would prefer bandlimited original signals too (they did not).
I don't think it is that simple although I 100 % support that observation .
I listen to what must be the most bandwidth limited material possible . Acoustic 78's . I still feel the bandwidth of the system needs to be large to play them correctly .
If 78's are corrected digitally that seems to be a different story . I think we might say bandwidth when we are collecting a number of related things together . Some are the poor relations of the family . I think I am right to say CD filtering and FM filtering have different outcomes ? One should expect a reconfigured FM filter ideal for 16 bit . I have no hands on experience so say as I read . FM filters for CD didn't work . Great apologies if wrong . Perhaps no one tried a filter in the 1960/70's style when CD ? That is similar was used but not as well executed ( cost ) .
Digital correction might still be very wide band even if configured as a filter . Might this matter ?
I listen to what must be the most bandwidth limited material possible . Acoustic 78's . I still feel the bandwidth of the system needs to be large to play them correctly .
If 78's are corrected digitally that seems to be a different story . I think we might say bandwidth when we are collecting a number of related things together . Some are the poor relations of the family . I think I am right to say CD filtering and FM filtering have different outcomes ? One should expect a reconfigured FM filter ideal for 16 bit . I have no hands on experience so say as I read . FM filters for CD didn't work . Great apologies if wrong . Perhaps no one tried a filter in the 1960/70's style when CD ? That is similar was used but not as well executed ( cost ) .
Digital correction might still be very wide band even if configured as a filter . Might this matter ?
60 years ago we were still not in the clear about quite a few things we know today. Nor did we have the semiconductors we have today.
From private e-maling with some of the members here, it is my understanding that we still do not KNOW exactly why is wide bandwidth (closed loop full power bandwidth) important, but we do know that somehow they almost always manage to sound at least a little bit better than limited bandwidth amps.
The question here is twofold. One, what is "wide bandwidth", how far does it need to go to be called that, and two, how was it obtained, by having a wide open loop bandwidth with little global NFB, or simply lots of global NFB.
Japanese brands during the 70ies sold their products as wide bandwidth if they hit say 300 kHz or above, but very little was known about exactly how they managed that. Spectral, on the other hand, seems unhappy with anything less than 1 MHz.
Personally, I agree completely with the Otala & Lohstroh approach. Make it work in full power open loop mode up to say 100 kHz, then apply say 20 dB of global NFB to clean up the output stage artefacts. They did it with off the shelf parts even then, 40 years ago, imagine what could be done today with modern parts.
It really makes me wonder.
Last edited:
What about Bruno's suggestion of more the merrier , of course open loop stability is paramount. I have issues with the upper midrange /top end of low bias, high NFB amplifiers, they tend to sound very unnatural, what is not unnatural is the speed and life they bring to the party (like good class-D) , loads of energy, but lots of hardness too...
Opposite of class-A ....
Opposite of class-A ....
That seems to be the usual trade-off.
Some years ago, I experimented with my own H/K integrated amps, models 6550 (50/70W 8/4 Ohms, SEPP) and 680 (85/130W 8/4 Ohms).
The 680 model uses two output device pairs per channel, and according to the service manual, each should be biased at 80 mA per trannie, 160 mA in total. I increased this by 10 mA in steps, and listened to it. Finally, I got to 140 mA per devices and it sounded no different to 130 mA, so I decreased it back to 130 mA.
The overall effect was that it made nice, warmer music, but it was less well defined overall.
To me it's very simple: it's necessary, so that FB can do its job properly - the actual action of the FB mechanism is that of a very high speed process, but if it can't function at those speeds then it most likely will do more harm than good. In one sense, the core of designing a good amplifier is making the inherent FB mechanism as competent, as 'accurate' as you possibly can ... then the output, as a reflection of the input, will take care of itself ...
We all know that FRank, but I believe there are aspects we do not yet fully understand.
As ever, nothing is that simple.
For example, we can all agree that having a full power OL bandwidth of 20 kHz is a good thing, since the amp will then not depend on NFB for normal operation. But, is 20 kHz enough? Should it be more, and if so, how much more? The wider it is, the harder it will be to stabilize the amp before applying global NFB.
What has a better chance of sounding good, an amp with an OL bandwidth of 100 kHz and 20 dB of NFB, or an amp with an OL bandwidth of 50 kHz and 26 dB of NFB?
As ever, nothing is that simple.
For example, we can all agree that having a full power OL bandwidth of 20 kHz is a good thing, since the amp will then not depend on NFB for normal operation. But, is 20 kHz enough? Should it be more, and if so, how much more? The wider it is, the harder it will be to stabilize the amp before applying global NFB.
What has a better chance of sounding good, an amp with an OL bandwidth of 100 kHz and 20 dB of NFB, or an amp with an OL bandwidth of 50 kHz and 26 dB of NFB?
Dvv . I analyzed your amplifier from the published diagram and found much to my surprise it can work . What you seem to have done is push the class A bit up to a point where it is us who distorts rather than being able to " accurately " detect a problem . It is a very well choosen point because you still have class B advantages of efficiency . Douglas Self goes to great trouble to suggest this is the worst design you can contemplate . This seems to be because it offends him . What I would call it is alternative class G . I know it isn't but it has the same compromises albeit much easier to build .
You degenerate your VAS which you do as a choice rather than because others have . I think this must make the amplifier more stable . Whereas most people build the VAS as an I to V converter yours looks to be a proper voltage amp . I wouldn't like to say how I define that ( too complicated even for me ) . The point is most people don't give a dam how linear that bit is as long as it does the job . That isn't so daft as the linearity is a function of loop feedback correction . The problem with local feedback ( it isn't strictly that if I to V ) is it robs loop feedback possibilities . VAS non linearity is usually second harmonic so not a nasty distortion . The VAS capacitance has a hidden advantage , it swaps the not so nice version inside the transistor . This is the main reason low Cob is preferred . Think about it , adding 27 pF makes low Cob a joke . Hitachi will say low Cob to enhance HF ability , how come ? Marginally I side with Hitachi as a nice NPO is better and is adjustable . People say not having a LTP input makes a VAS cap less important . No it doesn't . If it does it is a small difference . All the single input amps I built with low Cob went unstable ( fire mostly ) . No writer , it is that 1964 VAS transistors were about > 50pF .
Up to a point adjustment of VAS Re might give lower 20 kHz distortion . If VAS capacitance ( 10 to 100 pF ) can be relaxed and local emitter feedback applied the lowest 20 kHz distortion might be had ( yes it isn't really local feedback if trans-conductance ) . This also makes the driving of the VAS easier . My guess would be 10R with a 8 mA VAS current . The VAS might then be 22 pF rather than 47 pF . It might also have distortion 3 dB lower at 20 kHz . 2 pole compensation decisively better if asking ( e.g . 220 + 27 pF 100R to output stage , Never tried equal and suspect better if to output stage rather than 0V rail , If 0V uses 2K2 ) .
Douglas Self will say that reducing loop feedback has no beneficial effect . The evidence seems to be distortion at 500 Hz shoots up from - 105 db to - 85dB . I think I might live with that . Show me a sub-woofer with - 40 dB please . At 10 kHz where it matters it is equal as the stability is the bigger factor .
Thus your amp swamps the problem it has made and works . What Mr Self has no reliable figure for is how we hear . My conjecture is 2 watts class A should do it . 5 would be perfect .
The Quad 303 does it the other way . Very low bias and very well controlled . The 303 is on paper a class A design if the distortion means anything . In truth it has a bit of an MP3 effect due to class B . For all that it works . As a 1965(67) design it is very good . It never asks the components do anything they can not do . It also exceeds human ability to say it is wrong . It won't kill a pair of JBL's as it's only weakness . It will sound dull if your hi fi sounds dull . It will sound bass light if your hi fi is bass light . It will sound boring if .....
My conjecture is to set the bias in the middle would ruin both designs ( 60 mA ) . I do understand all NPN and complimentary differ in how bias might work , your amp is complimentary . In my heart I still think all NPN is worthy .
Mr Dvv , that is the how and the why of your amp . We two could never agree on that if building our own . I suspect if I still had a shop I would ask to sell your amp . You defend it too well to be wrong .
You degenerate your VAS which you do as a choice rather than because others have . I think this must make the amplifier more stable . Whereas most people build the VAS as an I to V converter yours looks to be a proper voltage amp . I wouldn't like to say how I define that ( too complicated even for me ) . The point is most people don't give a dam how linear that bit is as long as it does the job . That isn't so daft as the linearity is a function of loop feedback correction . The problem with local feedback ( it isn't strictly that if I to V ) is it robs loop feedback possibilities . VAS non linearity is usually second harmonic so not a nasty distortion . The VAS capacitance has a hidden advantage , it swaps the not so nice version inside the transistor . This is the main reason low Cob is preferred . Think about it , adding 27 pF makes low Cob a joke . Hitachi will say low Cob to enhance HF ability , how come ? Marginally I side with Hitachi as a nice NPO is better and is adjustable . People say not having a LTP input makes a VAS cap less important . No it doesn't . If it does it is a small difference . All the single input amps I built with low Cob went unstable ( fire mostly ) . No writer , it is that 1964 VAS transistors were about > 50pF .
Up to a point adjustment of VAS Re might give lower 20 kHz distortion . If VAS capacitance ( 10 to 100 pF ) can be relaxed and local emitter feedback applied the lowest 20 kHz distortion might be had ( yes it isn't really local feedback if trans-conductance ) . This also makes the driving of the VAS easier . My guess would be 10R with a 8 mA VAS current . The VAS might then be 22 pF rather than 47 pF . It might also have distortion 3 dB lower at 20 kHz . 2 pole compensation decisively better if asking ( e.g . 220 + 27 pF 100R to output stage , Never tried equal and suspect better if to output stage rather than 0V rail , If 0V uses 2K2 ) .
Douglas Self will say that reducing loop feedback has no beneficial effect . The evidence seems to be distortion at 500 Hz shoots up from - 105 db to - 85dB . I think I might live with that . Show me a sub-woofer with - 40 dB please . At 10 kHz where it matters it is equal as the stability is the bigger factor .
Thus your amp swamps the problem it has made and works . What Mr Self has no reliable figure for is how we hear . My conjecture is 2 watts class A should do it . 5 would be perfect .
The Quad 303 does it the other way . Very low bias and very well controlled . The 303 is on paper a class A design if the distortion means anything . In truth it has a bit of an MP3 effect due to class B . For all that it works . As a 1965(67) design it is very good . It never asks the components do anything they can not do . It also exceeds human ability to say it is wrong . It won't kill a pair of JBL's as it's only weakness . It will sound dull if your hi fi sounds dull . It will sound bass light if your hi fi is bass light . It will sound boring if .....
My conjecture is to set the bias in the middle would ruin both designs ( 60 mA ) . I do understand all NPN and complimentary differ in how bias might work , your amp is complimentary . In my heart I still think all NPN is worthy .
Mr Dvv , that is the how and the why of your amp . We two could never agree on that if building our own . I suspect if I still had a shop I would ask to sell your amp . You defend it too well to be wrong .
Having made the headphone amp I am very bored by it , a startling discovery was made . Some specs first .
Gain = 3
Distortion < 0.01% at any output level up to 20 kHz . 0.006% typical ( up to 10 kHz ) .
SE Class A .
200 kHz full power bandwidth
1 Hz - 3dB .
As a final test I added it to my mobile phone . At last a test of something I have never been able to do before . The outcome is interesting . It does not sound the same as the input by any stretch of the imagination . It is slightly bland and sounds more accurate . As this is a near zero distortion device I have to side with Bob Stewart and say an " accurate copy " of the input will not sound the same . As much as I despise headphones they do prove it . Instinctively it is right yet so hard to prove . I think why such illogic survives is saying the truth ruins the possibility of making a living . Engineers never get time to do better designs as the market moves quickly . They hide behind their Audio Precision analyzers and it does the listening for them ( I have the use of one when I need it , two in fact ) . I asked my friend John if he has an Audio Precision at work to which he said similar only better . He works in scientific research . He says with very little adjustment it will and does do that . He slightly conned the people where he works to build a massive power amp rather than buy an Amcron . He smiles and says how well it might work at home . I rather regret giving him a D Self book as it looks suspicious like it is a Blameless design .
I can see a number of flaws in my approach ( I need to add 32 R load to input ) . I will test it further and see where it goes . What is valuable here is a source rich in error . This is a signature . The headphone amp gives a fake signature . Lets be clear the amp is excellent . It is not identical and should be . It is not adding to the already rich spectrum of sounds . It is subtracting in the same way as any analogue copy .
Gain = 3
Distortion < 0.01% at any output level up to 20 kHz . 0.006% typical ( up to 10 kHz ) .
SE Class A .
200 kHz full power bandwidth
1 Hz - 3dB .
As a final test I added it to my mobile phone . At last a test of something I have never been able to do before . The outcome is interesting . It does not sound the same as the input by any stretch of the imagination . It is slightly bland and sounds more accurate . As this is a near zero distortion device I have to side with Bob Stewart and say an " accurate copy " of the input will not sound the same . As much as I despise headphones they do prove it . Instinctively it is right yet so hard to prove . I think why such illogic survives is saying the truth ruins the possibility of making a living . Engineers never get time to do better designs as the market moves quickly . They hide behind their Audio Precision analyzers and it does the listening for them ( I have the use of one when I need it , two in fact ) . I asked my friend John if he has an Audio Precision at work to which he said similar only better . He works in scientific research . He says with very little adjustment it will and does do that . He slightly conned the people where he works to build a massive power amp rather than buy an Amcron . He smiles and says how well it might work at home . I rather regret giving him a D Self book as it looks suspicious like it is a Blameless design .
I can see a number of flaws in my approach ( I need to add 32 R load to input ) . I will test it further and see where it goes . What is valuable here is a source rich in error . This is a signature . The headphone amp gives a fake signature . Lets be clear the amp is excellent . It is not identical and should be . It is not adding to the already rich spectrum of sounds . It is subtracting in the same way as any analogue copy .
Ooooooh, time for a field day! 😀
I see. You didn't expect me to be capable of it, hence your surprise. 😛 😛 😛
It would be helpful if you told WHICH schematic you refer to; frankly, I have no idea which one you are areferring to, there were several. I ASSUME you refer to Wayne's 1 Ohm Load Current factory.
Whichever, I did what I thought was best on basis of personal experience. It's no secret that I think Otala & Lohstroh were bull's eye to the point, what they said and did makes a lot of sense to me. It stands to reason.
Douglas Self's views do not stand to reason as I see reason. I am by no means discounting or belittling Mr Self, it's just that I disagree with him on some points. Not a question of right or wrong, a question of what one believes in.
I try, when possible, to judge a man's work by his results. I was told Mr Self was one of the key designers of the new Cambridge Audio products. I have heard only two amps, and frankly, I was not impressed. On basis of that, I think of Mr Self's work as of no interest to me. I have heard one line premp as published by Pavel Macura (PMA here) and it sounded really good to me, hence for me, PMA knows what he's doing and he's doing it well. I have heard only one power amp by John Curl, and I loved it, unfortunately it was snatched before I could buy it - hence, for me, John Curl knows what it's all about, and I happen to agree with his views to an almost frightening degree. It's that simple.
I believe, right or wrong, that it's the input stage which gives or doesn't most of what we call character of sound (assuming I got everything else right enough not to be problematic). If anything, I tend to overdo the input stage, like using FET terminated cascoded CCS, active (rather than passive) CCS for the input cascode, I tend to sprinkle capacitors all over, etc. In general, I will got to pains, possibly irrational pains, to decouple whatever I can from the power supply lines, and I am not stingy with them either.
Nige, as you well know, getting something to work electrically is just a good beginning. Getting it to sound great is the knack of it, and development is supposed to do just that, in my view, take a perfectly functional circuit and make it sound good.
In my view, one should park the bias of the VAS somewhere between 10 and 12 mA for many reasons, but the key one is that most of the likely candidates among transistors will by then have reached, or have gone very near to their maximum perofmance. Also, that is cpmfortably above the peak current the signal will need, meaning the trannies should use up say twice the worst case current they might need for the signal. Keeps them well in class A and keeps them linear.
I think (but am not sure) that Mr Self is far too concerned with the commercial side of things, so he needs low THD numbers, he needs specs. He's still too concerned with how it will look on paper.
I don't give a damn how it will look on paper because I know that if proper care is taken. it won't look bad at all.
The most important lesson of my audio life was delivered by the Otala & Lohstroh amp. Yes, I did make it in the mid 70ies. Its specs were just so-so regarding THD, but its sound made it very clear that THD specs had nothing to do with the sound delivered. That amp simply blew the competition away, despite being low powered for many. It made its point very clear. I think I got the point, but the proof is in the pudding.
Well, that depends on other factors as well, and heavily so. For someone with speaker efficiency of say 95 dB/2.83V/1m, even 1W in pure class A would probably be just fine. But for another, with speaker efficiency of say 88 dB/2.83V/1m, I'd say 3W would be a reasonable minimum. And so forth.
Then it depends on what one likes to listen to. If one listens mostly to quite chamber music, that's one thing, but if one is into thunderous heavy metal, that's an altogether different matter.
[qote]The Quad 303 does it the other way . Very low bias and very well controlled . The 303 is on paper a class A design if the distortion means anything . In truth it has a bit of an MP3 effect due to class B . For all that it works . As a 1965(67) design it is very good . It never asks the components do anything they can not do . It also exceeds human ability to say it is wrong . It won't kill a pair of JBL's as it's only weakness . It will sound dull if your hi fi sounds dull . It will sound bass light if your hi fi is bass light . It will sound boring if ..... [/quote]
Agreed, I like the 33 WAY better than the 405. I believe the 33 was made for music, while the 405 was made to prove a point, and despite my great respect for P. Walker and Quad, I think he missed the point. It works fine if your speakers are straight 8 Ohms, but if they fall below 6 Ohms, it starts to choke. Not so the 33.
Why, thank you, Mr Pearson, I am flattered (at least a bit). But I must say that if you're talking about what I think you are, the 1 Ohm Current Factory I did for Wayne, that is a topology that's been with me ever since the mid 70ies. I had all the time to think about it to grasp it as best I can.
As for disagreeing, my dear Mr Pearson, you know how we both LOVE to disagree one with another, but on the other hand, take a step back and cosider the results of that, for both of us. We both did learn and pick up a few points here and there from it, making our disagreement constructive and useful - as they should be.
Perhaps it's better that you don't have the shop any more, I'm not sure I am up to another manufacturing oddisey. What seems to be a hell of a problem in talking to Westeners is that they simply do not because they cannot uderstand the circumstances surrounding me. I do not have a background industry to fall back on, just making a simple aluminium case it took me literally years to resolve, and I am still dissatisfied with the quality obtained. Finding such mundane things like 2SC3503/2SA1381 trannies was a big problem for me, until recently. I literally have to order EVERYTHING from abroad, and that tends to build up the price no end.
I think my manufacturing days are over, but I can still play with one-off samples. Or three-off, no problem for anyone to chip in for PCBs, or some such.
I see. You didn't expect me to be capable of it, hence your surprise. 😛 😛 😛
It would be helpful if you told WHICH schematic you refer to; frankly, I have no idea which one you are areferring to, there were several. I ASSUME you refer to Wayne's 1 Ohm Load Current factory.
Whichever, I did what I thought was best on basis of personal experience. It's no secret that I think Otala & Lohstroh were bull's eye to the point, what they said and did makes a lot of sense to me. It stands to reason.
Douglas Self's views do not stand to reason as I see reason. I am by no means discounting or belittling Mr Self, it's just that I disagree with him on some points. Not a question of right or wrong, a question of what one believes in.
I try, when possible, to judge a man's work by his results. I was told Mr Self was one of the key designers of the new Cambridge Audio products. I have heard only two amps, and frankly, I was not impressed. On basis of that, I think of Mr Self's work as of no interest to me. I have heard one line premp as published by Pavel Macura (PMA here) and it sounded really good to me, hence for me, PMA knows what he's doing and he's doing it well. I have heard only one power amp by John Curl, and I loved it, unfortunately it was snatched before I could buy it - hence, for me, John Curl knows what it's all about, and I happen to agree with his views to an almost frightening degree. It's that simple.
I believe, right or wrong, that it's the input stage which gives or doesn't most of what we call character of sound (assuming I got everything else right enough not to be problematic). If anything, I tend to overdo the input stage, like using FET terminated cascoded CCS, active (rather than passive) CCS for the input cascode, I tend to sprinkle capacitors all over, etc. In general, I will got to pains, possibly irrational pains, to decouple whatever I can from the power supply lines, and I am not stingy with them either.
Nige, as you well know, getting something to work electrically is just a good beginning. Getting it to sound great is the knack of it, and development is supposed to do just that, in my view, take a perfectly functional circuit and make it sound good.
In my view, one should park the bias of the VAS somewhere between 10 and 12 mA for many reasons, but the key one is that most of the likely candidates among transistors will by then have reached, or have gone very near to their maximum perofmance. Also, that is cpmfortably above the peak current the signal will need, meaning the trannies should use up say twice the worst case current they might need for the signal. Keeps them well in class A and keeps them linear.
I think (but am not sure) that Mr Self is far too concerned with the commercial side of things, so he needs low THD numbers, he needs specs. He's still too concerned with how it will look on paper.
I don't give a damn how it will look on paper because I know that if proper care is taken. it won't look bad at all.
The most important lesson of my audio life was delivered by the Otala & Lohstroh amp. Yes, I did make it in the mid 70ies. Its specs were just so-so regarding THD, but its sound made it very clear that THD specs had nothing to do with the sound delivered. That amp simply blew the competition away, despite being low powered for many. It made its point very clear. I think I got the point, but the proof is in the pudding.
Well, that depends on other factors as well, and heavily so. For someone with speaker efficiency of say 95 dB/2.83V/1m, even 1W in pure class A would probably be just fine. But for another, with speaker efficiency of say 88 dB/2.83V/1m, I'd say 3W would be a reasonable minimum. And so forth.
Then it depends on what one likes to listen to. If one listens mostly to quite chamber music, that's one thing, but if one is into thunderous heavy metal, that's an altogether different matter.
[qote]The Quad 303 does it the other way . Very low bias and very well controlled . The 303 is on paper a class A design if the distortion means anything . In truth it has a bit of an MP3 effect due to class B . For all that it works . As a 1965(67) design it is very good . It never asks the components do anything they can not do . It also exceeds human ability to say it is wrong . It won't kill a pair of JBL's as it's only weakness . It will sound dull if your hi fi sounds dull . It will sound bass light if your hi fi is bass light . It will sound boring if ..... [/quote]
Agreed, I like the 33 WAY better than the 405. I believe the 33 was made for music, while the 405 was made to prove a point, and despite my great respect for P. Walker and Quad, I think he missed the point. It works fine if your speakers are straight 8 Ohms, but if they fall below 6 Ohms, it starts to choke. Not so the 33.
Why, thank you, Mr Pearson, I am flattered (at least a bit). But I must say that if you're talking about what I think you are, the 1 Ohm Current Factory I did for Wayne, that is a topology that's been with me ever since the mid 70ies. I had all the time to think about it to grasp it as best I can.
As for disagreeing, my dear Mr Pearson, you know how we both LOVE to disagree one with another, but on the other hand, take a step back and cosider the results of that, for both of us. We both did learn and pick up a few points here and there from it, making our disagreement constructive and useful - as they should be.
Perhaps it's better that you don't have the shop any more, I'm not sure I am up to another manufacturing oddisey. What seems to be a hell of a problem in talking to Westeners is that they simply do not because they cannot uderstand the circumstances surrounding me. I do not have a background industry to fall back on, just making a simple aluminium case it took me literally years to resolve, and I am still dissatisfied with the quality obtained. Finding such mundane things like 2SC3503/2SA1381 trannies was a big problem for me, until recently. I literally have to order EVERYTHING from abroad, and that tends to build up the price no end.
I think my manufacturing days are over, but I can still play with one-off samples. Or three-off, no problem for anyone to chip in for PCBs, or some such.
Last edited:
😎🙂
[Too bad, dvv, that you couldnt go live in a more parts/manufacturing friendly area of the planet.]
-Richard Marsh
[Too bad, dvv, that you couldnt go live in a more parts/manufacturing friendly area of the planet.]
-Richard Marsh
Nigel I am trying to get a feel for what it is you are comparing. What is 'the input' that sounds different? Headphone different to mobile phone, or to speakers?
Jan
Just the simplest test . The amp is mostly a buffer . A gain of 3 is my tentative attempt using CD as an input ( 2 V rms typical ) .
The output is a copy . It doesn't sound like one .
As the input is K ohms I guess my phone might be more in class A than is typical . This might explain the feeling that the sound is more accurate ( in fact rather good ) . The point is the phone can drive headphones . If all the verve is coming from crossover distortion that is a lesson for me . To my ears the buffer is less good . I am not heart broken . It is logical but hard to prove .
I understand what you are saying, Richard, but overall, I am not unhappy here where I am. Sure, most things are all wrong, but the few that matter to me are just as they should be. Anyway, most of the things wrong here are not all that better than where you live, although your sociatey is incomparably better organized and efficient than mine is.
Basically, I am a hard core family man, so I am happiest where my family is, and if they are all right, then so am I. And while I may lack much technology in terms of parts, actually I have all I need.
But, life can really have a joke at our expense. Many absolute first class cases several Ultra de Luxe Californian audio companies use are actually made by a countryman of mine, who emigrated in the late 80ies. Unfortunately, the prices of cases exceed the prices of my finished products, although they do look divine.
So if I get it, the headphones on the mobile sound different from the headphones on the headphone amp?
What't the source when you listen to the headphone amp, your mobile?
jan
Dvv . Most of your amps I have seen follow a pattern . I think all I said applies . Mostly what you seem to have changed is house keep circuits . As you know my idea of house keeping is a resistor if I can .
There is a saying if you ask an Irishman how to get to so and so he will say " I wouldn't start from here " It turns out this is no joke as it is impolite to ask as you are saying where you are is less good . The French say exactly the same and for sure the Italians . So the surprise is " I wouldn't start from here " . I looked at your general idea and said " it works " .
I am Dyslexic . It seems I use that memory for circuits . One famous designer who sometimes writes on DIY showed me an amp . He said " I can't give you it " . To which I said " you just did " .
Sorry if so much of what I write is less than my best ideas . It is to keep discussions going so as to get real opinions . I am in a world where I don't always understand what the world wants . The only way I have changed over the years is build to the desires of others . That sometimes means almost being them . What my designs have is mid band purity and verve . I call it the Caruso effect .
I think I first entered this thread with bullet proof wood used in German trains . I was making hi fi parts aged 4 . My dad showed me and when desperate to hear music I remembered what he did . I have a very bad soldering iron burn aged 8 .
I still remember his exact lecture about not opening the amplifier aged 4 ( I had and it was running ) . Instead of don't it was EXACTLY how it worked . The last bit was , touch that one if you want to be dead . His job was radar instructor . He is 83 now and remembers most of it . Baritone singer when he could be bothered . We owned the piano of Ketleby who wrote In a Persian Market ( everyone knows it , he got rich ,the music is at best amusing ) , his originals were in the piano stool . That's how I started loving this industry .
There is a saying if you ask an Irishman how to get to so and so he will say " I wouldn't start from here " It turns out this is no joke as it is impolite to ask as you are saying where you are is less good . The French say exactly the same and for sure the Italians . So the surprise is " I wouldn't start from here " . I looked at your general idea and said " it works " .
I am Dyslexic . It seems I use that memory for circuits . One famous designer who sometimes writes on DIY showed me an amp . He said " I can't give you it " . To which I said " you just did " .
Sorry if so much of what I write is less than my best ideas . It is to keep discussions going so as to get real opinions . I am in a world where I don't always understand what the world wants . The only way I have changed over the years is build to the desires of others . That sometimes means almost being them . What my designs have is mid band purity and verve . I call it the Caruso effect .
I think I first entered this thread with bullet proof wood used in German trains . I was making hi fi parts aged 4 . My dad showed me and when desperate to hear music I remembered what he did . I have a very bad soldering iron burn aged 8 .
I still remember his exact lecture about not opening the amplifier aged 4 ( I had and it was running ) . Instead of don't it was EXACTLY how it worked . The last bit was , touch that one if you want to be dead . His job was radar instructor . He is 83 now and remembers most of it . Baritone singer when he could be bothered . We owned the piano of Ketleby who wrote In a Persian Market ( everyone knows it , he got rich ,the music is at best amusing ) , his originals were in the piano stool . That's how I started loving this industry .
I have a little (tiny) Fiio headphone amp.
When inserted in series between my phone (or MP3/wav) player and headphones the sound is indeed different......same, but different.
NwAvGuy has lots to say about headphone amps.
Output impedance is important when driving headphones.
Nigel, perhaps this is what you are hearing ?.
Dan.
Dan I think it is exactly that and your description is correct . It is getting everything right so as to have covered all angles . I am not unhappy.
Same but different and actually getting somewhere .
The output impedance is interesting . It is about zero if I choose .
Same but different and actually getting somewhere .
The output impedance is interesting . It is about zero if I choose .
- Status
- Not open for further replies.