What are the general impressions of OPA2365.
I have a China made DAC in which they used AD8397 for the headphone OUT. Headphone output stage is basically a Cmoy operating at single supply 3.5 V. It sounds fine, just slightly better than my HTC Desire 820 Dual SIM. I ordered OPA2365 to replace the AD8397 there and possibly I will need extra supply bypass due to OPA2365's lower PSRR. If someone has general impressions on it's sound I would like to know.
As for bypass, has anyone tried [C+R||C]? It should provide a second order bypass, possibly better HF performance.
I have a China made DAC in which they used AD8397 for the headphone OUT. Headphone output stage is basically a Cmoy operating at single supply 3.5 V. It sounds fine, just slightly better than my HTC Desire 820 Dual SIM. I ordered OPA2365 to replace the AD8397 there and possibly I will need extra supply bypass due to OPA2365's lower PSRR. If someone has general impressions on it's sound I would like to know.
As for bypass, has anyone tried [C+R||C]? It should provide a second order bypass, possibly better HF performance.
OPA2365 is a CMOS opamp. Whilst I've not heard it, all the CMOS opamps I have heard haven't sounded much good at all in normal configurations. I'd be interested in others' experiences with them. I got AD8532 working OK but only with its supplies bootstrapped and fairly heavily bypassed with at least 1000uF low ESR.
abraxalito, what you said is true in general, but not in the case of OPA2365. OPA2365 has only a PMOS input pair thus has no so called crossover distortion. To achieve the rail-to-rail input range without the NMOS input pair, the supply to the input devices is raised beyond the external power supply.
Choosing a suitable design output from Wolfson 8741 or AK4399
I noted your replies Hurtig with interest and wondered if you could advise on design of an output from Wolfson 8741 or AK4399 DAC - so far the advice that I have been given is to output directly into a 600:600 transformer instead of an opamp - I am returning to electronics as a hobby after 30 years, so, am getting back 'up to speed' with a really enquiring mind...
I know quite a few op-amps
Have been working in audio R&D some years!
I know that op-amps sounds different. But still, you always know the sound of an op-amp.
I know to many people, who has been upgrading their gear with new op-amps, and been happy about... Just until they tried to put back the NE5532 in
And yes, I know it hasn't got the best specs in the world. But it has a very raw sonic performance No extra glance, extra softness and so on. What you put in, you will get out.
But if you realy do search the ultimate sonic performance, you should cinsider a discrete design, instead of the easy op-amp solution.
I noted your replies Hurtig with interest and wondered if you could advise on design of an output from Wolfson 8741 or AK4399 DAC - so far the advice that I have been given is to output directly into a 600:600 transformer instead of an opamp - I am returning to electronics as a hobby after 30 years, so, am getting back 'up to speed' with a really enquiring mind...
What Hurtig said, for the ultimate sonics, go discrete. But the DAC chips you mention already contain opamps and opamps tend to sound best when driving the lightest possible loads.
Hence I'd recommend not a direct connection to a transformer (at both the LF and HFends the trafo's quite possibly a fairly demanding load for an opamp, SQ-wise) rather use a discrete transistor buffer then the trafo.
Hence I'd recommend not a direct connection to a transformer (at both the LF and HFends the trafo's quite possibly a fairly demanding load for an opamp, SQ-wise) rather use a discrete transistor buffer then the trafo.
I'd just call them "good"
Also, depending on the use/replacement, make sure there's good local decoupling. (100 nF ceramic right at the pins)
Unfortunately, one important detail is not mentioned:
The application.
I think that making a comparison in a line level with a small gain factor or in a phono pre-amp makes a significant difference, which Op-Amp is to prefer.
I am currently interested in which dual op amps would be to prefer in a MM phono preamp with active RIAA equalization according the circuit topology in the attached images.
Thorsten Loesch prefer the OPA637 - in the iPhono2 the quad version
OV-4627A, go to
iFi Audio Model "micro iPhono2": Creating Schematic - help needed
and
ADA 4627 (ADA4627-1) outperform Burr Brown's OPA 627 (OPA627)
and Jeff Rowland prefer the AD744 resp. the dual version AD746 in his replacement phono modules of his preamplifier Coherence ONE
In most applications from commercial RIAA preamp devices according the attached images there is in use the BA4558, NE5532, OPA2134 and OPA2604.
By listening tests in preamp line stage with gain factor arround 4-6 best results I observe by the use with Op-Amps, where internal only one gain stage and high quiecent current in the output stage at the same time was realized.
Excellent examples are the AD797 (I want to know the dual version so as a version with additional jFETs at the input), AD846 and AD844.
Also good (but only with short signal lead to the power amplifier) are the AD829, AD825, AD847, OP604, AD8022 and AD8065/AD8066 (also single gain stage by use of a folded cascode + buffer stage, unfortunately low idle current through the output)
Not good for me are OPA627/637, NE5534, TDA1034, AD8597 and TL071 (probably because the use of two gain stages, i. e. LTP + VAS and the necessary compensation, but unfortunately within the second gain stage = VAS.
check out also post #26 under
What is wrong with op-amps?
Now for me rises up the above mentioned question, because I don't know, whether the same rules apply to the RIAA preamp application from the attachment or not.
Thank you for advices.
The application.
I think that making a comparison in a line level with a small gain factor or in a phono pre-amp makes a significant difference, which Op-Amp is to prefer.
I am currently interested in which dual op amps would be to prefer in a MM phono preamp with active RIAA equalization according the circuit topology in the attached images.
Thorsten Loesch prefer the OPA637 - in the iPhono2 the quad version
OV-4627A, go to
iFi Audio Model "micro iPhono2": Creating Schematic - help needed
and
ADA 4627 (ADA4627-1) outperform Burr Brown's OPA 627 (OPA627)
and Jeff Rowland prefer the AD744 resp. the dual version AD746 in his replacement phono modules of his preamplifier Coherence ONE
In most applications from commercial RIAA preamp devices according the attached images there is in use the BA4558, NE5532, OPA2134 and OPA2604.
By listening tests in preamp line stage with gain factor arround 4-6 best results I observe by the use with Op-Amps, where internal only one gain stage and high quiecent current in the output stage at the same time was realized.
Excellent examples are the AD797 (I want to know the dual version so as a version with additional jFETs at the input), AD846 and AD844.
Also good (but only with short signal lead to the power amplifier) are the AD829, AD825, AD847, OP604, AD8022 and AD8065/AD8066 (also single gain stage by use of a folded cascode + buffer stage, unfortunately low idle current through the output)
Not good for me are OPA627/637, NE5534, TDA1034, AD8597 and TL071 (probably because the use of two gain stages, i. e. LTP + VAS and the necessary compensation, but unfortunately within the second gain stage = VAS.
check out also post #26 under
What is wrong with op-amps?
Now for me rises up the above mentioned question, because I don't know, whether the same rules apply to the RIAA preamp application from the attachment or not.
Thank you for advices.
Attachments
Last edited:
If an opamp sounds neutral, there is a possibility that there is a short between input and output.
