Re: ERRONYMUS BOSCH.
....the purpose, and indeed, the beauty of negative feedback of course, is that such 'errors', non-linearity, noise,...etc...are rendered inaudible in a well designed amp...
fdegrove said:Hi,
It is still an error no matter how small.
Thanks to compounded small errors we miss out on fidelity which is why we should consider the error in every single component, heck, even an inch of wire or a silly fuseholder for all I care.
Cheers,
....the purpose, and indeed, the beauty of negative feedback of course, is that such 'errors', non-linearity, noise,...etc...are rendered inaudible in a well designed amp...
NOPE...
Hi,
No Sir, the errors are there and always will be.
What NFB does is attenuate them by a defined amount usually expressed in dB and by the same token introduce other timing related problems.
The only beauty I find in feedback is when it is applied locally, not globally and even then I remain skeptical of it.
Feeding back,
Hi,
No Sir, the errors are there and always will be.
What NFB does is attenuate them by a defined amount usually expressed in dB and by the same token introduce other timing related problems.
The only beauty I find in feedback is when it is applied locally, not globally and even then I remain skeptical of it.
Feeding back,
Re: Really?
Been a bit under the weather.
Now I'm not, I'm going to take my bike for a spin through the mountains.
Cheers
fdegrove said:
Been a bit under the weather.
Now I'm not, I'm going to take my bike for a spin through the mountains.
Cheers
Re: Re: Re: ERRONYMUS BOSCH.
...correct..beautifully written thesis not withstanding....see cordell....'A MOSFET Power Amplifier with error correction', and refereces cited theroff, with regard to DIM, TIM, etc, and relationship to THD.....Note that the author of the thesis cites a zero feedback design 'cary, reviewed in stereophile, where the reviewer asserts the amp comminicates i n a way he has not experianced before....is this not the same publication that refered to the Halcro, (with substantial feedback, and error correction), as the worlds best amp.?......consistency is not the province of these publications i fear...
i take issue with this thesis from its very title...which implies that existing methods of measurement do 'not correlate' with perceived sound quality......
Brett said:
...correct..beautifully written thesis not withstanding....see cordell....'A MOSFET Power Amplifier with error correction', and refereces cited theroff, with regard to DIM, TIM, etc, and relationship to THD.....Note that the author of the thesis cites a zero feedback design 'cary, reviewed in stereophile, where the reviewer asserts the amp comminicates i n a way he has not experianced before....is this not the same publication that refered to the Halcro, (with substantial feedback, and error correction), as the worlds best amp.?......consistency is not the province of these publications i fear...
i take issue with this thesis from its very title...which implies that existing methods of measurement do 'not correlate' with perceived sound quality......
Re: NOPE...
...What 'timing related' problems does feedback introduce?
...and on what grounds do you consider local negative feedback to be superior to its global application?
...and precisely why do you nevertheless remain sceptical of the whole shooting match...
...do'nt get me wrong, if there is something you are attempting to teach me here, i am more than willing to learn , but please be gentle.... ....i need to be taken from first principals....
fdegrove said:Hi,
No Sir, the errors are there and always will be.
What NFB does is attenuate them by a defined amount usually expressed in dB and by the same token introduce other timing related problems.
The only beauty I find in feedback is when it is applied locally, not globally and even then I remain skeptical of it.
Feeding back,
...What 'timing related' problems does feedback introduce?
...and on what grounds do you consider local negative feedback to be superior to its global application?
...and precisely why do you nevertheless remain sceptical of the whole shooting match...
...do'nt get me wrong, if there is something you are attempting to teach me here, i am more than willing to learn , but please be gentle....
....i need to be taken from first principals....
Re: Harold S. Black.
That rather describes our whole existence, doesn't it?
I mean, everything we experience we experience after the fact. Light. Sound. Never impulses. They all travel at finite speeds. We can only experience, respond and react to our surroundings as they existed in the near past, never in the present.
se
fdegrove said:
That rather describes our whole existence, doesn't it?
I mean, everything we experience we experience after the fact. Light. Sound. Never impulses. They all travel at finite speeds. We can only experience, respond and react to our surroundings as they existed in the near past, never in the present.
se
Re: I'M IMPRESSED.
How does it show that I don't grasp the fact that feeback in amplifiers is an event after the actual time event, time delayed after the natrual time delay? If I hadn't grasped that fact, I wouldn't have expanded it to include our own sensory existence. I didn't say anything specifically about negative feedback in amplifiers because there was simply nothing to add.
Oh, and I'm still waiting to hear your reasoning behind 100kHz being a "limited frequency response" in the context of audio.
se
fdegrove said:
How does it show that I don't grasp the fact that feeback in amplifiers is an event after the actual time event, time delayed after the natrual time delay? If I hadn't grasped that fact, I wouldn't have expanded it to include our own sensory existence. I didn't say anything specifically about negative feedback in amplifiers because there was simply nothing to add.
Oh, and I'm still waiting to hear your reasoning behind 100kHz being a "limited frequency response" in the context of audio.
se
Re: Harold S. Black.
A common enough misconception...ie...delayed feedback....can never occur in a stable system,(can anybody see why?), if it did you would have a very efficient power oscillator......
fdegrove said:Hi,
Error correction, no matter how fast, is always, by nature, a correction occuring after the event, in casu an error.
Hence my skepticicsm,
A common enough misconception...ie...delayed feedback....can never occur in a stable system,(can anybody see why?), if it did you would have a very efficient power oscillator......
Re: Re: Harold S. Black.
Well let's see, the only way to NOT have delayed feedback would be if you happened to live in some strange parallel universe where electromagnetic waves propagated with infinite speed. On this planet, you can only ideally do as good as the speed of light, which is near infinitely slower than infinite.
So if no delay of feedback can occur in a stable system, then there must not be a stable system anywhere on earth.
se
mikek said:
Well let's see, the only way to NOT have delayed feedback would be if you happened to live in some strange parallel universe where electromagnetic waves propagated with infinite speed. On this planet, you can only ideally do as good as the speed of light, which is near infinitely slower than infinite.
So if no delay of feedback can occur in a stable system, then there must not be a stable system anywhere on earth.
se
Re: Re: Re: Harold S. Black.
.....you cannot have delayed feedback for a stable negative feedback audio power amp..........the output, and by implication and the fraction thereof that is fed back, must respond instanteneously to input stimuli, even though it may, in charging and discharging design and parasitic time constants, take longer to complete it's response....
Steve Eddy said:
.....you cannot have delayed feedback for a stable negative feedback audio power amp..........the output, and by implication and the fraction thereof that is fed back, must respond instanteneously to input stimuli, even though it may, in charging and discharging design and parasitic time constants, take longer to complete it's response....
Re: Re: Re: Re: Re: Harold S. Black.
Yup. Unless you live in mikek's parallel universe where electromagnetic waves propagate with infinite speed.
In this universe, the best you can ideally achieve is 300 million meters per second. In realworld applications, a percentage of that depending on the medium which the electromagnetic waves are propagating.
se
planet10 said:
Yup. Unless you live in mikek's parallel universe where electromagnetic waves propagate with infinite speed.
In this universe, the best you can ideally achieve is 300 million meters per second. In realworld applications, a percentage of that depending on the medium which the electromagnetic waves are propagating.
se
Re: Re: Re: Re: Re: Harold S. Black.
http://diyaudio.com/forums/showthread.php?postid=150727#post150727
Basically, the feedback "speed" is a non-issue that appears to have been created for the reasons in the post referenced above.
Null testing with music driving real speakers is sufficient proof for anyone who wants to verify that high nfb amplifiers, as a rule, indeed have fewer kinds of ALL distortions than low NFB amps do.
I think what mikek was trying to say is if you create even a fraction of a cycle of feedback delay at ultrasonic frequencies--somewhere around 0.000005 seconds--you end up with an oscillator instead of an amplifier. As Steve Eddy correctly points out, the speed at which the signal moves through copper is so fast as to simply not be an issue here.
Quoting Self, p43:
"No linear circuit can introduce a pure time-delay; the output must begin to respond at once, even if it take a long time to complete its response. In the typical amplifier the dominant-pole capacitor introduces a 90 degree phase shift between the input-pair and the output at all but the lowest audio frequencies, but this is not a true time delay. The phrase 'delayed feedback' is often used to describe this situation, and it is a wretchedly inaccurate term. If you really delay the feedback to a power amplifier, (which can only be done by adding a time-constant to the feedback network rather than the forward path) it will quickly turn into the proverbial power oscillator as sure as night follows day."
So fdegrove was simply raising a common audiophile red herring here with no basis in fact.
You might want to see my post here:planet10 said:
http://diyaudio.com/forums/showthread.php?postid=150727#post150727
Basically, the feedback "speed" is a non-issue that appears to have been created for the reasons in the post referenced above.
Null testing with music driving real speakers is sufficient proof for anyone who wants to verify that high nfb amplifiers, as a rule, indeed have fewer kinds of ALL distortions than low NFB amps do.
I think what mikek was trying to say is if you create even a fraction of a cycle of feedback delay at ultrasonic frequencies--somewhere around 0.000005 seconds--you end up with an oscillator instead of an amplifier. As Steve Eddy correctly points out, the speed at which the signal moves through copper is so fast as to simply not be an issue here.
Quoting Self, p43:
"No linear circuit can introduce a pure time-delay; the output must begin to respond at once, even if it take a long time to complete its response. In the typical amplifier the dominant-pole capacitor introduces a 90 degree phase shift between the input-pair and the output at all but the lowest audio frequencies, but this is not a true time delay. The phrase 'delayed feedback' is often used to describe this situation, and it is a wretchedly inaccurate term. If you really delay the feedback to a power amplifier, (which can only be done by adding a time-constant to the feedback network rather than the forward path) it will quickly turn into the proverbial power oscillator as sure as night follows day."
So fdegrove was simply raising a common audiophile red herring here with no basis in fact.
Steve Eddy said:
I understand everything about DBT, but you wouldn't say that it is common practice to have DBT listening intervals longer than few minutes. I have never participated in DBT that lasted a week
Steve Eddy said:
Of course; switching should be as fast as possible, but listening intervals shouldn't be short.
To me, analogy with taste (or other human senses) is obvious;
if I take a sip of wine (or beer) I'm not very familiar with, and than take sip of another similar sort of wine (and every other parameter is the same - temperature, glass etc.) I'm sure I wouldn't be able to tell the difference. But if somebody switch a beer I'm drinking for a long period of time for some other brand I would probably notice a difference.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.