Page 2...this took longer than I'd expected. But here is a case in point!
Mikes' support of Andyc's request in another thread.
Fair shout.
Then...
All fire and brimstone and belief. Not a shred of explanation. Put in a wrapper of pretentious circumlocution (that's "jargon" to the rest of us).
Mike, to be credible and add value you must adapt your style to fit your audience. Otherwise you cannot expect your self-respecting audience to be receptive. I'm not sure whether Carlos is being serious or sarastic but I would not make the mistake, Mike, of believing that people react unfavourably to you because you are some sort of mis-understood genius. You certainly aren't that!
To me you are like some sort of tyrannical librarian: you collect seemingly endless references to all sorts of exotic hyperbole and then ram it down unsuspecting peoples' throats. It seems you are either oblivious to how to communicate effectively to your audience or you are deliberately diminishing others as part of some bizarre self-aggradizement.
Having ideas is worth nothing on its own. Everyone has ideas, lots and lots of them. Value is in what you do with those ideas, specifically how those ideas influence and impact other people in a positive manner. To make an impact you must be persuasive. The best persuaders explain themselves to the satisfaction of their audience (not themselves) and do so in an unassuming manner.
You are smart enough to comprehend this. Are you smart enough to see the benefit of it and act on it? I think you are.
Mikes' support of Andyc's request in another thread.
Fair shout.
Then...
All fire and brimstone and belief. Not a shred of explanation. Put in a wrapper of pretentious circumlocution (that's "jargon" to the rest of us).
Mike, to be credible and add value you must adapt your style to fit your audience. Otherwise you cannot expect your self-respecting audience to be receptive. I'm not sure whether Carlos is being serious or sarastic but I would not make the mistake, Mike, of believing that people react unfavourably to you because you are some sort of mis-understood genius. You certainly aren't that!
To me you are like some sort of tyrannical librarian: you collect seemingly endless references to all sorts of exotic hyperbole and then ram it down unsuspecting peoples' throats. It seems you are either oblivious to how to communicate effectively to your audience or you are deliberately diminishing others as part of some bizarre self-aggradizement.
Having ideas is worth nothing on its own. Everyone has ideas, lots and lots of them. Value is in what you do with those ideas, specifically how those ideas influence and impact other people in a positive manner. To make an impact you must be persuasive. The best persuaders explain themselves to the satisfaction of their audience (not themselves) and do so in an unassuming manner.
You are smart enough to comprehend this. Are you smart enough to see the benefit of it and act on it? I think you are.
traderbam said:
In other words.....to reduce all forms of distortion, wrap as much feedback around as simple, and as inherently linear a signal path as possible....
This is evident in such excellent designs as the AD797, NE5532/3/4,
OPA632...etc
Moreover, there is no evidence...anywhere that feedback is inherently bad for an amplifiers health.....
This, Traderbam, is what gets my goat.......
traderbam said:
No???.....are you sure??
http://diyaudio.com/forums/showthread.php?postid=483377#post483377
traderbam said:Having ideas is worth nothing on its own.
Everyone has ideas, lots and lots of them.
Value is in what you do with those ideas, specifically how those ideas influence and impact other people in a positive manner.
To make an impact you must be persuasive.
The best persuaders explain themselves to the satisfaction of their audience (not themselves) and do so in an unassuming manner.
You are smart enough to comprehend this.
Are you smart enough to see the benefit of it and act on it? I think you are.
How very English....!
No...seriously...point almost entirely inapplicable, but well made nonetheless....Cheers.
Traderbam, that is very well written. English is not my native language, and I could never express myself in such a clear way. So I have to wait that another one does this job
May I add: I hope you are
And may I address that to others (they know who they are) in this forum too.
For the benefit of the forum.
Too much very knowledgeable people left already, and I have the impression that it is because of such fights about THEORY.
This is sad for those like me (=very few knowledge in electronics, no experience which cirquits SOUND well and which not) glad to participate to the EXPERIENCE of those guys.
Some of them developed a lot of well sounding stuff (to my ears, and obviously not only to mine), so from my perspective, they are enough credible. I'll never have so much time to build and listen to a lot of circuits. So what's easier than listen what these guys say, and follow their advise?
(He he, as long as it is not their intention to lead us in the wrong direction-joke)
Ciao, Tino
traderbam said:You are smart enough to comprehend this. Are you smart enough to see the benefit of it and act on it? I think you are.
May I add: I hope you are
And may I address that to others (they know who they are) in this forum too.
For the benefit of the forum.
Too much very knowledgeable people left already, and I have the impression that it is because of such fights about THEORY.
This is sad for those like me (=very few knowledge in electronics, no experience which cirquits SOUND well and which not) glad to participate to the EXPERIENCE of those guys.
Some of them developed a lot of well sounding stuff (to my ears, and obviously not only to mine), so from my perspective, they are enough credible. I'll never have so much time to build and listen to a lot of circuits. So what's easier than listen what these guys say, and follow their advise?
(He he, as long as it is not their intention to lead us in the wrong direction-joke)
Ciao, Tino
Andy C, my interest in open loop bandwidth was re-ignited when I read the paper by Barrie Gilbert on PIM as well as other problems with op amp type designs.
When I hear that someone is completely pleased with a 5534 IC op amp, it makes me laugh.
When this design first came out, I knew that it met the TIM criterion, but what would it sound like? Well, I built a test comparator between an IC op amp and one of my discrete designs. It was for a phono stage, and I SWAPPED the RIAA components between the two designs, so that the SAME eq was provided. I spoke about this in 1979 or so, in 'The Audio Amateur' in my discussion with Dr. Lipshitz. I then measured the differences between the two circuits and found them to be very small. However, in listening, the IC op amp circuit sounded too 'smooth', like it was removing detail. I then knew that we had NOT solved the problem with audio problems with negative feedback, just by building a somewhat better IC.
Barrie's paper apparently has been removed from the internet. We might have to contact him directly in order to get a clean copy.
I can't find a copy at the moment, but when Barrie Gilbert actually admits to a problem with op amps, I listen! You should too!
When I hear that someone is completely pleased with a 5534 IC op amp, it makes me laugh.
When this design first came out, I knew that it met the TIM criterion, but what would it sound like? Well, I built a test comparator between an IC op amp and one of my discrete designs. It was for a phono stage, and I SWAPPED the RIAA components between the two designs, so that the SAME eq was provided. I spoke about this in 1979 or so, in 'The Audio Amateur' in my discussion with Dr. Lipshitz. I then measured the differences between the two circuits and found them to be very small. However, in listening, the IC op amp circuit sounded too 'smooth', like it was removing detail. I then knew that we had NOT solved the problem with audio problems with negative feedback, just by building a somewhat better IC.
Barrie's paper apparently has been removed from the internet. We might have to contact him directly in order to get a clean copy.
I can't find a copy at the moment, but when Barrie Gilbert actually admits to a problem with op amps, I listen! You should too!
john curl said:
I am afraid that especially this would strongly depend on one's taste, what will be prefered. And the "detail removing" might be very questionable. How about other characteristics, like clarity and resolution of complex orchestral music?
P.S. I do not vote for NE5534/5532, as I greatly dislike the sound of those opamps.
I read that article by Barrie as well, and thought it was very good. I remember he mentioned Otala in there as well. I think that EDN removed it from their site. Unfortunately I didn't save a local copy.
I don't remember all the intimate details, but I am quite sure he did not mention high open-loop bandwidth as a requirement for eliminating the AM to PM conversion. I could be wrong of course, but him having said that would surprise me very much. The reason I say this is that I remember basically agreeing with everything in there, and if he had said that wide open-loop bandwidth was a requirement, I know I would have remembered disagreeing with that.
I'm not arguing that all op-amps are good in this regard, or that Barriie's analysis was incorrect. I do take issue with Matti's conclusion about the open-loop bandwidth, but I'm not one of those that has thrown out the baby with the bathwater in that regard. The main thing I wanted to discuss, providing you were interested in such discussion (you haven't said yet), was the open-loop bandwidth issue. I just ordered the AES 20-CD set last night (nudge nudge, say n'more, say n'more), so I want to read Matti's PIM article when it arrives. I read his TIM article long ago, as well as Marshall's articles and Walt's EDN article which seems to have an overview of the PIM issue.
Or maybe this is just a communication issue related to bandwidth vs gain-bandwidth product?
I don't remember all the intimate details, but I am quite sure he did not mention high open-loop bandwidth as a requirement for eliminating the AM to PM conversion. I could be wrong of course, but him having said that would surprise me very much. The reason I say this is that I remember basically agreeing with everything in there, and if he had said that wide open-loop bandwidth was a requirement, I know I would have remembered disagreeing with that.
I'm not arguing that all op-amps are good in this regard, or that Barriie's analysis was incorrect. I do take issue with Matti's conclusion about the open-loop bandwidth, but I'm not one of those that has thrown out the baby with the bathwater in that regard. The main thing I wanted to discuss, providing you were interested in such discussion (you haven't said yet), was the open-loop bandwidth issue. I just ordered the AES 20-CD set last night (nudge nudge, say n'more, say n'more), so I want to read Matti's PIM article when it arrives. I read his TIM article long ago, as well as Marshall's articles and Walt's EDN article which seems to have an overview of the PIM issue.
Or maybe this is just a communication issue related to bandwidth vs gain-bandwidth product?
I think we should not condemn opamps, as it makes no sense. Every opamp is an unique circuit design and they behave different. I would advocate for opamps with first pole of Aol above audio band. This is achievable with CFB opamps, and they do sound considerably better than VFB opamps. I think that the biggest problem of a conventional opamp is frequency roll-off -6dB/oct of Aol starting at very low freguencies.
john curl said:
If there is something wrong with the NE5532/3/4, it most certainly has nothing to do with its open-loop bandwidth....
The only conceivable problem as far as i am aware, is the generation of a slow settling component in it's transient response.
This is due to the feed-foward path around the second stage giving rise to imperfect pole-zero cancellation (viz: doublet).
The feedfoward path is necessary for stability in this three-gain stage design....
Settling time is an issue in conversion systems, but i fail to see how it can be a significant problem in phono preamps.....
In any case i gave a raft of examples, including AD797 and OPA632 to illustrate the point.......
andy_c said:
Andy, virtually all of the TIM work discussed by Otala, Leach et al, really pertains to single-pole, miller compensated topologies, for which the concept of constant gain-bandwidth product pertains...
Change compensation to double-pole, or multi-pole nested comp., and constant- gain bandwidth goes out the window in a hurry....
stuff like SID......TIM can then be persuaded to disappear in the audio band...
Folks, let me reply to all of you, in general. Anyone who thinks that a 5534 IC is 'good enough' for hi end audio products, makes me laugh! I seriously tested it over 25 years ago, and found it lacking.
The AD797 is a much better op amp, but not perfect for all audio applications. However, I compete with designs using the AD797, and I think it is one of the best IC op amps available. Still, I will stick to discrete designs for my best efforts. Why? Because I can use class A EVERYWHERE in preamp designs, and complementary jfets anywhere I want. I like jfets, they are VERY quiet, have a high impedance input, and have a more linear transfer function. IC's don't have complementary jfets, it isn't practical at this time to put them in. Therefore, Nelson, Charles, and I can do interesting and sophisticated circuits that are not possible with IC's.
Then, there is the problem with thermal feedback. ALL IC'S have thermal feedback, some more than others. This might be MORE important than open loop bandwidth, who knows.
I have the Barrie Gilbert article around here somewhere, but I can't find it just now. I will re-read the section with PIM being discussed, when I find it. However, I find most feedback by Mikek and AndyC not what I remember reading. Let's find out what the article really said.
The AD797 is a much better op amp, but not perfect for all audio applications. However, I compete with designs using the AD797, and I think it is one of the best IC op amps available. Still, I will stick to discrete designs for my best efforts. Why? Because I can use class A EVERYWHERE in preamp designs, and complementary jfets anywhere I want. I like jfets, they are VERY quiet, have a high impedance input, and have a more linear transfer function. IC's don't have complementary jfets, it isn't practical at this time to put them in. Therefore, Nelson, Charles, and I can do interesting and sophisticated circuits that are not possible with IC's.
Then, there is the problem with thermal feedback. ALL IC'S have thermal feedback, some more than others. This might be MORE important than open loop bandwidth, who knows.
I have the Barrie Gilbert article around here somewhere, but I can't find it just now. I will re-read the section with PIM being discussed, when I find it. However, I find most feedback by Mikek and AndyC not what I remember reading. Let's find out what the article really said.
Yes but how do you explain this to someone who just added NFB to their circuit and lost much of the music? In this sense feedback is bad: it has made things worse. The real problem is that the understanding of how feedback works is missing. In colleges feedback is almost always taught using perfectly linear systems. People seem to forget that semiconductors are non-linear and expect feedback to achieve the same results. I happen to believe that feedback is such a cunning and hugely beneficial technique that it is remis not to make use of it. But you have to really understand what it is to be able to use it beneficially.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.