I already made it clear at the beginning that I was doing this for the challenge.
Not with the intension to beat any IC solutions, especially in spec.
Whatever you prefer or think is scientific better is fine with me.
I have a lot of fun doing this, which is what matters.
And some other people think it is worth the trouble too :
https://www.diyaudio.com/forums/pas...s-diy-headphone-amp-guide-28.html#post5780536
(#1375)
Cheers,
Patrick
Not with the intension to beat any IC solutions, especially in spec.
Whatever you prefer or think is scientific better is fine with me.
I have a lot of fun doing this, which is what matters.
And some other people think it is worth the trouble too :
https://www.diyaudio.com/forums/pas...s-diy-headphone-amp-guide-28.html#post5780536
(#1375)
Cheers,
Patrick
I have no doubt about it, i might be wrong, but you'd probably better go back to Gray and Sedra for this effort to be better accomplished.
The biggest problem i see about this effort is that most of us here seem to be biased by Diyaudio topics which are on repeat mode most of the time.
We have about a few thousand pages on op-amps good for audio on this forum and Google can hardly find a damn thing about m5220,m5219, njm2043 and njm2041 which were used in most high end cassette players of the 90's and in early but very high quality I/V sections in early cd players that are still venerated, plus in countless good quality mixers, but people concentrated on ne5534 for some reason...njm2043 and njm2114 can beat ne5534 in any application. Even the "horrible" lm833 can beat ne5534 in i/v stages. Thank God for NOT listening to Diyaudio as they still sell the much discredited lm833.
Well... the best op amps were not made for audio, they were expensive and used in medical , metrology or military technology.
You would have INA117 or SSM2017 at 10 dollars each or at 100 dollars each in a different package with different specs.
The order of performance magnitude between the audio op-amps and those op amps developed by Analog Devices, Burr Brown, Intersil , Harris , Hitachi, Nec and others was simply high.
The most underrated spec was the thermal dissipation which actually improve by large the op-amp behavior cause even if you have some identical specs for both versions, like the industrial vs military chip, those specs are guaranteed either at 25 degrees Celsius or at 150 degrees Celsius and the difference is enormous for fast and powerful transients to be processed cleanly.
Now look at these two op-amps datasheets that i attached and, please, tell me what do you think about the audio op-amp plethora.
The datasheet tells that they are around at least since 1996...If i wouldn't log on Tapeheads i would've never knew this op-amp existed.
I asked for a few times here if anybody knows the reason why Denon and Nakamichi used in its most expensive players and dacs upc4570 in the i/v stage and ne5534 as a buffer after that...It clearly wasn't for the fact that upc was cheaper...they used two damn tarnsformers and a ton of very high quality capacitors and regulators plus a magnetic levitation transport in those models.
Do you see any talk about upc4570 qualities? The datasheet itself isn't very transparent but Denon knew something.
We still compare the whole op-amps world with ne5534 as the past and the new op-amp jewels of the present as the only solution that can only be beaten by dicrete op-amps.It's simply biased behavior.
The biggest problem i see about this effort is that most of us here seem to be biased by Diyaudio topics which are on repeat mode most of the time.
We have about a few thousand pages on op-amps good for audio on this forum and Google can hardly find a damn thing about m5220,m5219, njm2043 and njm2041 which were used in most high end cassette players of the 90's and in early but very high quality I/V sections in early cd players that are still venerated, plus in countless good quality mixers, but people concentrated on ne5534 for some reason...njm2043 and njm2114 can beat ne5534 in any application. Even the "horrible" lm833 can beat ne5534 in i/v stages. Thank God for NOT listening to Diyaudio as they still sell the much discredited lm833.
Well... the best op amps were not made for audio, they were expensive and used in medical , metrology or military technology.
You would have INA117 or SSM2017 at 10 dollars each or at 100 dollars each in a different package with different specs.
The order of performance magnitude between the audio op-amps and those op amps developed by Analog Devices, Burr Brown, Intersil , Harris , Hitachi, Nec and others was simply high.
The most underrated spec was the thermal dissipation which actually improve by large the op-amp behavior cause even if you have some identical specs for both versions, like the industrial vs military chip, those specs are guaranteed either at 25 degrees Celsius or at 150 degrees Celsius and the difference is enormous for fast and powerful transients to be processed cleanly.
Now look at these two op-amps datasheets that i attached and, please, tell me what do you think about the audio op-amp plethora.
The datasheet tells that they are around at least since 1996...If i wouldn't log on Tapeheads i would've never knew this op-amp existed.
I asked for a few times here if anybody knows the reason why Denon and Nakamichi used in its most expensive players and dacs upc4570 in the i/v stage and ne5534 as a buffer after that...It clearly wasn't for the fact that upc was cheaper...they used two damn tarnsformers and a ton of very high quality capacitors and regulators plus a magnetic levitation transport in those models.
Do you see any talk about upc4570 qualities? The datasheet itself isn't very transparent but Denon knew something.
We still compare the whole op-amps world with ne5534 as the past and the new op-amp jewels of the present as the only solution that can only be beaten by dicrete op-amps.It's simply biased behavior.
Last edited:
And many of us have fun reading about it, I think it's pretty cool. Thanks for posting your efforts.
I am not alone. 
https://www.diyaudio.com/forums/the...owtorch-preamplifier-iii-359.html#post5776050
https://www.diyaudio.com/forums/the...owtorch-preamplifier-iii-359.html#post5775952
(post #17914, 17905)
Patrick
https://www.diyaudio.com/forums/the...owtorch-preamplifier-iii-359.html#post5776050
https://www.diyaudio.com/forums/the...owtorch-preamplifier-iii-359.html#post5775952
(post #17914, 17905)
Patrick
There are a completely clear reason - total lack of knowledge.
No one engineer wants real competition in his working area.
Who knows about - works and doesn't present there, who present there - couldn't cope with modern opamps loopgain.
Pro Discrete Op Amp SS2590 - SparkoS Labs Inc.
https://www.diyaudio.com/forums/vendor-s-bazaar/329463-hypex-nc500-buffer-upgrade.html#post5787486
https://sparkoslabs.com/wp-content/uploads/2018/03/SS2590.pdf
looking at the datasheet i can only be amused...it's only the power and the power supply that's better than any modern audio op-amp, nothing else...
https://www.diyaudio.com/forums/vendor-s-bazaar/329463-hypex-nc500-buffer-upgrade.html#post5787486
https://sparkoslabs.com/wp-content/uploads/2018/03/SS2590.pdf
looking at the datasheet i can only be amused...it's only the power and the power supply that's better than any modern audio op-amp, nothing else...
Last edited:
But surely the profit margin ($/unit) is massively higher? A very important performance criterion for many! And the self-congratulation coefficient is pretty high too
Obviously lacking is the current noise density, so there's no way to judge its noise figure in most small signal circuitry. Its apparently optimized for voltage noise with the paralleled input devices, which suggests it is not optimized for current noise (more important in reality for most uses, since typical circuitry is at 1k to 20k impedance, not 10 to 300 ohms!) The (uncorrelated?) input bias current cancellation circuitry doesn't bode well for this either.
The circuit looks good, the high gain, EF VAS and two-pole compensation should give great THD, but they never actually state the THD or graph it, which comes across as suspicious - it ought to perform well but why keep schtum in that case? The class AB output should give an interesting compromise in performance for heavy loads - which will cause an increase in output stage distortion without affecting the performance at low currents. Its basically a good BJT power amp with the feedback network and output stage chopped off AFAICT. The lack of THD measurements is immediately curious - perhaps something goes badly wrong with the output biasing?
Googling didn't seem to throw up anything about the current noise or measured THD or IM.
A simple power buffer added to OPA454 (6 euros each for 1 pcs)and there's no competition in that voltage range...
http://www.ti.com/lit/ds/symlink/opa454.pdf
http://www.ti.com/lit/ds/symlink/opa454.pdf
"Cascode structures inside of the device will begin to debias when the outupt swing approaches 3.5V
away from either supply rail. This is non destructive, and the output will not visibly clip, however
there will be a ~25dB reduction in open loop gain, resulting in higher THD."
This is terrible.
Voltage noise density 35nV, 300nV at 10Hz. But its FET inputs so current noise not a problem. The compromises made for high voltage are costly - the dynamic range is much less than a standard part due to the voltage noise. Flicker noise knee seems to be somewhere between 1kHz and 10kHz.
The THD at x10 gain is about 25 times more than at unity gain, and rises precipitously above 1kHz to 0.035%
The GBW product is 3MHz, explaining the bad HF distortion at gain rather well.
So doesn't stack up at all against regular parts at +/-17V since all of the extra headroom is pissed away on noise, and the basic bandwidth is missing.
It's not a single pole amplifier, so the effective feedback isn't as simple as determining the 0dB gain frequency (3MHz) and counting backward at 6dB per octave. For example, the amplifier open loop gain shown on the graph is about 85dB at 10kHz, not possible with a 3MHz single pole amplifier.
I guess this also means that there are circuits you can put it in that it will not be stable, but it'll probably work over some sort of gain range - have not explored this amp in detail.
I guess this also means that there are circuits you can put it in that it will not be stable, but it'll probably work over some sort of gain range - have not explored this amp in detail.
Last edited:
For the majority of applications where you're working with the "usual" rail voltages that's pretty poor headroom.
Overall there are better, cheaper choices that would offer better performance, even at higher rail voltages.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.