Dear all,
As the initiator of this post it is nice to see all the replies although as sometimes happens the subject moves in another direction compared to the original question.
By now I see I have some options but the single Op-Amp per channel (discrete or IC
) should be fine in my design.
Unless some of the newer posters have some comments on that ... I am all ears ...
Oystein
As the initiator of this post it is nice to see all the replies although as sometimes happens the subject moves in another direction compared to the original question.
By now I see I have some options but the single Op-Amp per channel (discrete or IC
) should be fine in my design.Unless some of the newer posters have some comments on that ... I am all ears ...
Oystein
As a new poster, I can comment that you have a choice of 2 versions (at least) of opamps designed for audio called 4562. But I have to warn you that if you decide to go with NJM4562 you can't make a voltage gain of the amp less than 10, otherwise it will oscillate on high frequency, while with LM4562 you can have a unity gain buffer.
Good luck!
Good luck!
KSTR said:Now that's a completely different part. Only the model number matches by pure coincidence. Still this was a nice sidetracking by Anatoliy
The original datasheet "Rev.E" is from 2003, the IC supposedly is much older.
- Klaus
Thanks. That's what I suspected. So Wavebourn's "4562" comments are regarding an entirely different part.
Back to your design, Oystein, nobody in this thread has demonstrated any evidence that any other part or discrete design is objectively superior to the LM4562 and several have agreed the LM4562 is the most transparent op-amp they know of. So hopefully that helps?
RocketScientist said:
Exactly the same confusion may happen with VN2222 MOSFET.
http://www.datasheetcatalog.org/datasheet/supertex/VN2224.pdf
http://www.datasheetcatalog.org/datasheet/calogic/VN2222.PDF
Confusion over part numbers isn't helpful or relevant to this discussion. What's relevant to Oystein is your statement:
I disagree with the above. For Oystein's application I think one can make a strong arguement the LM4562 (or one the newer LME versions from the same op-amp family) is indeed "optimal". Instead of trying to take the thread off topic, it would be helpful if you presented factual, objective and relevant evidence why the National LM4562 is "suboptimal" for Oystein's application? Do you have a better realistic solution, even ignoring cost and size, for his application you can prove outperforms the LM4562?
Wavebourn said:
I disagree with the above. For Oystein's application I think one can make a strong arguement the LM4562 (or one the newer LME versions from the same op-amp family) is indeed "optimal". Instead of trying to take the thread off topic, it would be helpful if you presented factual, objective and relevant evidence why the National LM4562 is "suboptimal" for Oystein's application? Do you have a better realistic solution, even ignoring cost and size, for his application you can prove outperforms the LM4562?
Sorry RocketScientist, as a novice in the audio and on the forum I can't present anything that sounds better than LM4562 opamp. I just expressed my humble opinion based on some little design experience and few listening tests. I believe it is a DIY forum, so amateurs like me have rights to express own opinions among RocketScientists. 
RocketScientist do you work for national or have any personal involvement or friendships with any one related to national ???.
Specs come second place regarding audio so Ill not get into this as I very rarely use chips, the odd AD826, AD829 or AD797. I havent tried the metal can national chips yet but I can say that I have tried the 8 pins in some mods with a couple of customers audio equipment and they didnt like it, and no it was not oscilating or anything like that.
You should rather say the audiophiles that national paid to review these chips liked them, dont forget there are many others that feel otherwise and from the looks of it theres quite a few and in the majority. I havent seen the national chips used in any top notch equipment this far, I have seen AD chips though but most opt for discretes. Its early days yet, well see how it goes in future.
Audio differencing is the exactly what I do, strangely I gather we get very different results. Well the only thing I can say in my defence is that I have no vested interest in any opamp company and evaluate purely on what my ears can hear. There will never be consensus corcerning these things but it would be interesting to see what well established and respected gurus like Nelson Pass, John Curl or Charles Hason would say about it.
Specs come second place regarding audio so Ill not get into this as I very rarely use chips, the odd AD826, AD829 or AD797. I havent tried the metal can national chips yet but I can say that I have tried the 8 pins in some mods with a couple of customers audio equipment and they didnt like it, and no it was not oscilating or anything like that.
You should rather say the audiophiles that national paid to review these chips liked them, dont forget there are many others that feel otherwise and from the looks of it theres quite a few and in the majority. I havent seen the national chips used in any top notch equipment this far, I have seen AD chips though but most opt for discretes. Its early days yet, well see how it goes in future.
Audio differencing is the exactly what I do, strangely I gather we get very different results. Well the only thing I can say in my defence is that I have no vested interest in any opamp company and evaluate purely on what my ears can hear. There will never be consensus corcerning these things but it would be interesting to see what well established and respected gurus like Nelson Pass, John Curl or Charles Hason would say about it.
I can also "just express my humble opinion based on some little design experience and few listening tests", I would rather use a discrete buffer stage than an op-amp. Some op-amps sound quite good but not one that I've tried, including the LM4562, could compare with the SQ of a discrete class A stage.
No worries Wavebourn. I'm fine with opinions--especially when it's a theoretical or philosophical discussion and there are no consequences involved. But when you factor in the DIY cost to spin PC boards, buy all the parts, all the time involved, etc. I try to steer fellow DIY'ers down the most ideal path possible. Most of us don't have unlimited time or budgets. And my personal preference, where possible, is to to offer suggestions I can back up with factual objective information and references.
Sadly, there are lots of myths perpetuated in high-end audio--some by very experienced individuals. Some myths have some truth behind them, some are pure fantasy, and with others we may never know. Those who say one solution is better or worse than another without providing any solid reasons just help perpetuate the myths despite many having little or no basis in fact. That's not sound engineering practice.
Personally, I hate to see someone go out and buy expensive op-amps, or hassle with a discrete design, when the outcome is likely to be no better--and more likely worse--than a much easier and less costly solution. And no, I have no affiliation in any way with National or current or former National employees beyond having a met a few of them at AES conferences, etc.
Sadly, there are lots of myths perpetuated in high-end audio--some by very experienced individuals. Some myths have some truth behind them, some are pure fantasy, and with others we may never know. Those who say one solution is better or worse than another without providing any solid reasons just help perpetuate the myths despite many having little or no basis in fact. That's not sound engineering practice.
Personally, I hate to see someone go out and buy expensive op-amps, or hassle with a discrete design, when the outcome is likely to be no better--and more likely worse--than a much easier and less costly solution. And no, I have no affiliation in any way with National or current or former National employees beyond having a met a few of them at AES conferences, etc.
homemodder said:
And I have no interest in National (see my previous post). If AD comes out with a better chip, I'll start using those
There are way too many favorable reviews of products using the LM4562 for National to have paid them off. A good example are the many favorable reviews of the Benchmark Dac1 Pre which uses the LM4562 exclusively throughout the device. It's been rave reviewed by even Stereophile and the UK mags, as well as by several recording engineering mags on the pro side. Some have called the Dac1 Pre the best sounding DAC they've ever heard. If the Dac1 doesn't qualify as "top notch" by your standards, I'm sorry, but I think in terms of specs it's yet to be beat.
I do agree there will never be overall consensus about these sorts of things. But I, at least, have lots of favorable numbers and reviews on my side.
RocketScientist said:
So, am I allowed to post?
Thank you!
Here is my humble opinion about how a buffer can be built:
http://www.diyaudio.com/forums/showthread.php?postid=1889739#post1889739
The problem is, I myself and lot of other humble amateurs presented on this forum don't believe that THD measurement is the final criterion for audio, especially when THD goes up with lower output voltage swing, what you may clearly see on graphs of National's datasheet. However, I don't blame on National for that. I just expressed my humble opinion that if costs and sizes are not involved in optimization equations, opamps are suboptimal for audio amplifications.
Wavebourn said:
As you hopefully know, the THD only rises because they're showing THD+N and, as the level drops, noise dominates the measurement rather than distortion. Regardless, even at extremely low levels, the THD+N of the LM4562 outperforms anything else I've ever seen used for audio. So it's a moot point.
And I agree THD is far from the only measurement of audio performance. The LM4562 excels at any measurement I'm aware of including transient tests, intermodulation distortion, noise, CMRR, PSRR, square waves, etc. And, ultimately, it also excels at playing real music as evidenced by audio difference testing and the many favorable listening reviews.
Wavebourn said:
I expected this objection, thank you.
Also, I forgot to stress that THD goes up with frequency. I myself and several humble amateurs believe that it is also a very bad, counter-clear-audio symptom.
Virtually any amplifier with feedback applied exhibits rising distortion with frequency--even a fully class A discrete design. Even most open loop (non-feedback) amps exhibit some rise in distortion with frequency.
If you don't believe THD is important or especially audible, than it shouldn't matter if it rises or falls? If you do think THD is important and audible (as you suggest above), than I think it's important to quantify how much THD there is at high frequencies?
Here again, the LM4562 will beat nearly anything else at any frequency you want to pick in terms of THD. So are you saying you'd rather have say an open loop class A design that at any frequency likely has literally 1000 times as much THD as the LM4562 but the high level of distortion rises less at high frequencies? If so, please explain how that is better?
And if you have an example of a very low distortion line-level audio amp that doesn't rise with frequency, please provide a reference?
RocketScientist said:
Dear Rocket Scientist;
I did not say that I don't believe that THD is important and especially audible; I just expressed my humble opinion that it is not the single final criterion that may be used to compare solutions. But it may be used as a symptom in case of absence of other facts. Also, I believe that there is one and only one way to reproduce clean sound, while number of ways to distort it is countless. That's why more and more of measurement methods are being invented, but they still can't cover all errors that may be created. It reminds me a crowd of persons that instead of going to the goal always try to avoid obstacles, so always have to back up bumping into walls, trees, poles, each other, falling in holes, and so on, because concentrating on problems (distortions), running from them backward, that crowd of smart persons don't see what is behind.
May I ask you please, to follow the link that I posted before? There I described the concept of a buffer that I used in a hybrid power amp, in a class A+C amp driver, and in line level output stages. I would appreciate if you express your educated scientific opinion about it.
Here it is, again:
http://www.diyaudio.com/forums/showthread.php?postid=1889739#post1889739
Actually, it is a source follower that is bootstrapped on top in order to keep a voltage between source and drain stable, non-modulated by an amplified signal. Also, it is loaded on current source that is modulated by an input signal, in counter - phase. As the result, the source follower exhibits the lowest possible distortions without any additional negative feedback that may create multiple distortions that were already caught, and other distortions that may be caught up in the future by scientists and engineers who are still curious why all known measurements don't correlate with listening tests, and what kind of errors must be measured to understand why they still don't correlate, instead of an usual excuse that people who claim to hear distortions are crazy.
Edit: "additional" word was added for strictness of expression of my humble opinion.
Wavebourn said:
I followed the link the first time, thanks. I don't consider it an example of anything Oystein might want to use in his pre-amp. And, trying to be polite, I didn't think you would appreciate my opinion of the "Tower-III" power amp so I thought it better to say nothing. I will, however, buy you a very nice dinner if the line stage has less distortion than an LM4762.
It appears to be a rather unique design--at least to my eyes. But it's not an example, in my opinion, of something superior to the LM4762--at least by any objective measure. And it's also a safe bet it has distortion that rises at high frequencies like virtually every other audio amplifier--including the $10,000 and up models so loved by the audiophile press.
If you honestly want more of an opinion, I'm probably not the best person to ask as I don't personally do much with tube or hybrid tube designs. I'm not a big fan of their almost always higher and often audible distortion--euphonic or otherwise.
But I also recognize lots of guys like how tubes sound and I respect that. Some of us like our audio signal as uncorrupted as possible and others like to pour ketchup on their $40 steak because it tastes better to them that way. It's ultimately about what puts the biggest smile on your face while listening to music.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.