Hi Bonsia,
Inductance within a closed NFB loop is not the same as having inductance between an output node and the LS system.
Inductance withing the loop can affect global loop performance and capabilities in the presence of reactive loads. It can trigger phantom non-linearity events which might arise at some particular instant of a music reproduction performance with a particular loudspeaker, but not others.
Cheers ......... Graham.
Inductance within a closed NFB loop is not the same as having inductance between an output node and the LS system.
Inductance withing the loop can affect global loop performance and capabilities in the presence of reactive loads. It can trigger phantom non-linearity events which might arise at some particular instant of a music reproduction performance with a particular loudspeaker, but not others.
Cheers ......... Graham.
Maybe there would have been less commotion if John had posted earlier what he's written in post #263 (to which I can agree).
I presume that Glen is mainly designing amps with the ubiqitous single-loops/dominant-pole compensation topology.
With these I would agree with him that some attempts in increasing stability into capacitive loads might degrade other parameters. But one is neither restricted to dominant-pole compensation nor single-loop topologies.
If only dominant compensation of control loops were known, then the ones who develop aircraft, guided missiles, industrial robots etc would be in deep trouble.
And the question remains what would be a reasonable capacitive load for amp testing. 2uF is a little much IMO.
Regards
Charles
I presume that Glen is mainly designing amps with the ubiqitous single-loops/dominant-pole compensation topology.
With these I would agree with him that some attempts in increasing stability into capacitive loads might degrade other parameters. But one is neither restricted to dominant-pole compensation nor single-loop topologies.
If only dominant compensation of control loops were known, then the ones who develop aircraft, guided missiles, industrial robots etc would be in deep trouble.
And the question remains what would be a reasonable capacitive load for amp testing. 2uF is a little much IMO.
Regards
Charles
Graham Maynard said:
Well, dear Graham, if I can help others to realise how unrealistic and totally irrelevant your input signal is, on which you base all your arguments, that's all the productivity I need right now.
Cheers, Jan Didden
Graham Maynard said:
No, honestly I don't. If you base a whole framework of arguments, studies, conclusions etc on a signal that is totally irrelevant, I don't see why any reasonably intelligent person comes to the conclusion that this should be keenly studied.
It would be SOO easy to convince us. Do your testing with audio-band signals. If the effect shows up, we would all fall over each other to want to learn more from your discovery. But, to be honest again, I'm not holding my breath.
And why do you refuse to label your graphs? This is elementary if you want to communicate. You really should read up on the scientific method of research and how to accumulate evidence that others will accept. Or, finding out that you are on the wrong track, which in itself is invaluable as well. Think of all the wasted time and effort if that would be the case but unknown!
Jan Didden
Graham Maynard said:
Correct. But, of course, the output coil is not within the fb loop. Don't start to mislead us on yet another issue.
Jan Didden
Graham Maynard said:
Hi Graham,
People, with or without 'flawed design skills' , generally will put the coil OUTSIDE the NFB loop.
Cheers,
Graham, maybe I misunderstand you, but for output coil is always outside the feedback loop - so I would not take the NFB point after the coil. If this is what you are objecting to, then yes, I'd say that's a risky thing to do.
Regarding the point you make about speakers, what I was trying to say is that I think the intrinisic sound differences between speakers is so great as to conmpletely swamp any effect of the output coil - whether the design uses a coil or not. From this vantage point, my conclusion is that the coil is not the key determining factor in an amplifiers performance or perceived sound. Sorry to keep repeating it, but surely we have to look at this as a system and then pareto the things with the parget probable impact on the sound. Would the coil be near the top of the list?
cheers
Regarding the point you make about speakers, what I was trying to say is that I think the intrinisic sound differences between speakers is so great as to conmpletely swamp any effect of the output coil - whether the design uses a coil or not. From this vantage point, my conclusion is that the coil is not the key determining factor in an amplifiers performance or perceived sound. Sorry to keep repeating it, but surely we have to look at this as a system and then pareto the things with the parget probable impact on the sound. Would the coil be near the top of the list?
cheers
Hilarious discussion
So we are discussing the ill effects of a 2uH inductor, which has no significance on the sonics at all. Have guys lost your thinking,how could be an inductor of such low value would degrade the sound of an amplifier, when other strong horrible factors are present in the vicinity of NFB, the back emf, impedance change in voice coil etc.
So we are discussing the ill effects of a 2uH inductor, which has no significance on the sonics at all. Have guys lost your thinking,how could be an inductor of such low value would degrade the sound of an amplifier, when other strong horrible factors are present in the vicinity of NFB, the back emf, impedance change in voice coil etc.
phase_accurate said:
My comments were about amplifier design in general, and power amplifier’s in general use dominant-pole compensation. Besides, the insinuation that they only apply to single-loop/dominant-pole compensated amplifiers is simply false. A capacitive load can throw a Hawksford error-corrected output stage out of stability or cause a loss of phase margin in a (for example) Cherry nested-differentiating-feedback amplifier or a feed-forward compensated amplifier as well. Attempting to avoid either case by compensating for excess phase margin into a purely resistive load will come at the expense of less high-frequency negative feedback.
zinsula said:
Now thats a strike...
So what are you doing in this thread??
If you have built several (different, I assume) amps and there is no audible difference, then it's either
1. You have built the ultimate amps, absolutely perfect.
2. Your listening abilities are not adequate
3. You are experiencing what many are accused of here: You hear what you want to hear, and as you are someone who seems to believe only in numbers, you'll not hear any difference, because all your designs are at 0.000....% distortion. Even if there is a difference.
In case 1, you should immediately go and sell these amps, they will sell like....
In case 2, what are you discussing here, with your listening abilities? You're not up to the task.
In case 3, you should show a little bit more respect towards the experienced designers here and stop winding them up.
In any case, I and many others are diyers here and would like to be able to continue to listen to JC, GM and all the others. Guys like you have already managed to drive away enough experienced people...
So, give it a rest.
Tino
Hmm.....
1) I never said that my SS amplifiers were/are absolutely perfect.
2) I do not believe that my SS amplifiers are absolutely perfect.
3) I think that whatever subtle sonic differences/colorations may possibly exist are perfectly negligible.
4) That’s why I refer to them as “transparent”.
5) I would call many amplifiers "transparent", not just my own.
6) I do not design my amps to only measure adequately in terms of distortion
7) I am not trying to wind anyone here up.
I don’t care about anyone’s ‘guru’ status and I’ll respect an opinion as far as it makes sense to me and as far as its proponent can back it up with a rational argument, and I sure as fudge don’t see what’s offensive about challenging some of the more dubious claims made here.
I’d really like for someone to tell me how they can hear the damped ‘ringing’ of a 2uH inductor in series with a capacitive load of a uF or two (whose series resonate frequency is well above the audio spectrum) for a start.
I mean, that has been alledged here.
Cheers,
Glen
Well, I must be completely incompetent and deaf to boot. People have paid me a lot of good money for amps with 7uH (yeah, seven microhenry!) output inductors. I did find the inductor had to be perfectly linear (air core, minimal outside influences) to avoid increasing the distortion within the audio band, but it was still a minor effect. The effects of the output inductor are measurable, but not audible to my ears under my listening conditions. The effects of an unstable amp are definitely audible, even to me. Theory's great, but it's hardware in the system and on the test bench that tells you how serious the problem is.
IMO, it may be difficult to make a practical loudspeaker load, other than a loudspeaker. I'd strongly suggest measuring the impedance and phase angle of a *real* speaker system in terms of G & B, and plotting that. Use a lot of resolution and don't forget the speaker cables. It's a lot more revealing than the usual presentation. You'll find it difficult to duplicate with any small combination of passive parts; owning a wire factory could also come in handy. Using that same model in a simulation, which fortunately places no limits on components, the magnitude of any effects at the amp output should be no mystery. It then comes back to the old question we can't answer- is it audible?
IMO, it may be difficult to make a practical loudspeaker load, other than a loudspeaker. I'd strongly suggest measuring the impedance and phase angle of a *real* speaker system in terms of G & B, and plotting that. Use a lot of resolution and don't forget the speaker cables. It's a lot more revealing than the usual presentation. You'll find it difficult to duplicate with any small combination of passive parts; owning a wire factory could also come in handy. Using that same model in a simulation, which fortunately places no limits on components, the magnitude of any effects at the amp output should be no mystery. It then comes back to the old question we can't answer- is it audible?
Graham Maynard said:
Graham,
Interestingly this last plot features group delay excursions from about -200 nS to 1700 nS in contrast with -1800 nS to 3900 nS, which may let us conclude this other box works better, but this is not really the point.
To argue that a few uHy of coil have a distinctly audible impact, taking into account the intrinsic unavoidable series inductance and other electrical and electromechanical reactances is ludicrous.
But of course I agree with Mr. Curl in that if something, no matter how negligible its effect may be, can be deleted, then it should be deleted. This makes perfect engineering sense.
It is usually the case some notions go unchallenged because there is widespread agreement as to whether they are important or not. In this case, any mainstream engineering will quickly agree a few uHy in series with the speaker cable and the horrendous reactive load a speaker is, are not significant at audio frequencies. A few quick numbers will support this.
It is also true thinking along the beaten path stiffles innovation or consolidates misconceptions, and it takes some mavericks challenging the statu quo to open new roads. But while this is a necessary condition - to think out of the accepted wisdom - it is not sufficient for innovation. For this to happen there must be some roadblock common wisdom cannot sort, a rationalization why this happens to be, and an alternate framework encompassing the old theory and the beakthrough discovery which clears the road.
Nothing of this has been exposed here convincingly, being the sole argument that somehow this is "obviously hearable".
Show me a double blind A/B test, or better and easier, simulate with and without output coil with actual music, and play back the difference. Show us how to repeat the experiment independently. Only then you will have a case, otherwise this may go on endlessly with no results.
Rodolfo
Ok lets Sign a "TREATY OF NO OUTPUT COILS" in amplifiers from Present to Future...... 😀 😀 😀 😀 😀 😉
Who would be the first one to signature....
Who would be the first one to signature....
I would sign without hesitation. Built amps without coil 30 years ago (just out of lazyness) and didn't have problems with them, but I didn't listen to 2 uF capacitors either !
Regards
Charles
Regards
Charles
phase_accurate said:
Charles is the first man in this world to sign this treaty[very daring] so, whose next..
I think John Curl would be the next one to get enlisted, ok get me listed also just for support.🙂
GUYS, This is becoming ABSOLUTELY RIDICULOUS.
Jan - Edmond - Bonsai - Jacco -
Shame on you all. Your imaginations are overly powerful.
Where have I mentioned having an inductance inside a NFB loop ?
PLEASE ANSWER !
Glen,
Who here defending the chokeless design has said that a 2uH choke rings ?
Hi Charles,
Your post#282 is on the button.
Hi Jan,
I can understand your first post#283 statement, but it is totally unsupported.
My Graph was labelled, so stop trying to score cheap points.
I had to cover it to squeeze within the diyAudio attachment size limit !!!
THAT IS WHY I DESCRIBED IT WITH TEXT FOR EVERYONE TO READ.
Anything wrong with that ?
Hi Paul,
No not 2uH in particular.
I said I heard effect of the standard 5 to 10uH values 35 years ago and so I would no longer countenance more than 0.5uH. Only now everyone cannot contain themselves, this has become a feeding frenzy of misrepresentation.
Hi Glen,
Don't you think it would have been better to mention post#289 before, instead of posturing when other contributors are trying to be open. I thought I might have missed something when I asked you to clarify, but I had not !
No one can be expected (should attempt to) second guess what anyone else might have in mind.
You also list things you did not say; seems to affect more than you !!!
That's a good post#292 Conrad,
at least you have stated clearly what you have and have not heard with your equipment.
Rodolfo,
Guys on this forum claim 0.01% THD at 20kHz and are so pleased at this. The error voltage introduced by a 5uH choke can be 100 times greater !!!
Yes indeed there is a distinct need for a dose of reality here.
Rodolfo - remenber - I cannot show you - I cannot present you with audible proof, which of course you know and thus you can make your challenge.
You need to try this for yourself as I have stated; only then might you understand.
Until choke/no choke usage has been tried definitively there will always be those who make challenge without substantiation.
Hi Workhorse,
You know you have my signature for chokeless design, but I fear it will do no more than start 'them' and 'us' cliques, like PP and SETs, etc.
Cheers ........ Graham.
Jan - Edmond - Bonsai - Jacco -
Shame on you all. Your imaginations are overly powerful.
Where have I mentioned having an inductance inside a NFB loop ?
PLEASE ANSWER !
Glen,
Who here defending the chokeless design has said that a 2uH choke rings ?
Hi Charles,
Your post#282 is on the button.
Hi Jan,
I can understand your first post#283 statement, but it is totally unsupported.
My Graph was labelled, so stop trying to score cheap points.
I had to cover it to squeeze within the diyAudio attachment size limit !!!
THAT IS WHY I DESCRIBED IT WITH TEXT FOR EVERYONE TO READ.
Anything wrong with that ?
Hi Paul,
No not 2uH in particular.
I said I heard effect of the standard 5 to 10uH values 35 years ago and so I would no longer countenance more than 0.5uH. Only now everyone cannot contain themselves, this has become a feeding frenzy of misrepresentation.
Hi Glen,
Don't you think it would have been better to mention post#289 before, instead of posturing when other contributors are trying to be open. I thought I might have missed something when I asked you to clarify, but I had not !
No one can be expected (should attempt to) second guess what anyone else might have in mind.
You also list things you did not say; seems to affect more than you !!!
That's a good post#292 Conrad,
at least you have stated clearly what you have and have not heard with your equipment.
Rodolfo,
Guys on this forum claim 0.01% THD at 20kHz and are so pleased at this. The error voltage introduced by a 5uH choke can be 100 times greater !!!
Yes indeed there is a distinct need for a dose of reality here.
Rodolfo - remenber - I cannot show you - I cannot present you with audible proof, which of course you know and thus you can make your challenge.
You need to try this for yourself as I have stated; only then might you understand.
Until choke/no choke usage has been tried definitively there will always be those who make challenge without substantiation.
Hi Workhorse,
You know you have my signature for chokeless design, but I fear it will do no more than start 'them' and 'us' cliques, like PP and SETs, etc.
Cheers ........ Graham.
- Status
- Not open for further replies.