I can detect the sound of the NEs, and I've proved it several times just by listening to a CDP I didn't knew.
When we opened the CDP, there they were.
In a word, the NEs sound DIGITAL.
Complete crap, lifeless, undynamic, spitty, sibillant.
Even a very old CDP I have (Telefunken
, no remote, LED display, mono 16-bit BB Dac with switching
before the op-amps) made two guys crazy comparing to their much more recent CDPs (one of them was a Sony SACD, the other was a Denon CDP).
I showed them the Telefunken because I'm bad, as I have much better.
The Telefunken has BB OPA2132 (well bypassed) on the analog stage.
What evoluted more: the Dacs or the op-amps?
You already know my oppinion.
When we opened the CDP, there they were.
In a word, the NEs sound DIGITAL.
Complete crap, lifeless, undynamic, spitty, sibillant.
Even a very old CDP I have (Telefunken
, no remote, LED display, mono 16-bit BB Dac with switching before the op-amps) made two guys crazy comparing to their much more recent CDPs (one of them was a Sony SACD, the other was a Denon CDP).I showed them the Telefunken because I'm bad, as I have much better.
The Telefunken
has BB OPA2132 (well bypassed) on the analog stage.What evoluted more: the Dacs or the op-amps?
You already know my oppinion.
In a word, the NEs sound DIGITAL.
Complete crap, lifeless, undynamic, spitty, sibillant.
-------------------------------------------------------------------
Agreed. At the suggestions of jhietbrink in Audioasylum, I tried the LM6172 with a 0.1 uF stacked NPO ceramic across V+ and V-. Previously, I never liked the 6172 but this time as filter/output from a Wolfson 192k chip, the sound is very good, clear with good timing and even lush! Much better than a biased AD826.
Complete crap, lifeless, undynamic, spitty, sibillant.
-------------------------------------------------------------------
Agreed. At the suggestions of jhietbrink in Audioasylum, I tried the LM6172 with a 0.1 uF stacked NPO ceramic across V+ and V-. Previously, I never liked the 6172 but this time as filter/output from a Wolfson 192k chip, the sound is very good, clear with good timing and even lush! Much better than a biased AD826.
Re: Impressively idiots
It is fairly naive to think that they have other considerations than numbers when they write their datasheets. None of us would have even looked twice at this DAC if it was specified with an opamp with 90dB SNR even though it might have a superior sound. Audio engineers and DIYaudio-people are not immune to marketing either.
If you want to achieve the exact same specifications as in the datasheet, then you implement the reference circuit. If you have other priorities then you are free to change this any way you want.
carlosfm said:
antness said:
It is fairly naive to think that they have other considerations than numbers when they write their datasheets. None of us would have even looked twice at this DAC if it was specified with an opamp with 90dB SNR even though it might have a superior sound. Audio engineers and DIYaudio-people are not immune to marketing either.
If you want to achieve the exact same specifications as in the datasheet, then you implement the reference circuit. If you have other priorities then you are free to change this any way you want.
carlosfm said:The LM6171 and LM6172 are amongst the best op-amps on Earth, IMHO.
I love'em.
I don't like the sound of the AD826, and even the AD825.
For me, they don't have midband.
Very recessed...
I must admit I (we) stopped using opamps years ago and
the 627 was about the last one and best at that time.
All the stuff I design now (from ground up) is discrete. However
there are still many times when an opamp is required for
upgrades or when cost is an issue.
I'm keen to try this LM6171, I know it works well for I-V due to
it's speed, however have you used it for other applications. Does
it drive cables/heavy loads well?
Cheers,
Terry
Terry,
Your reply to my previous post was both refreshingly polite and objective.
The amplifier is a modified JLH class-A, details of which can be found at... http://www.tcaas.btinternet.co.uk/jlhupdate.htm
The speakers are (recently completely rebuilt) Quad ESL-57s, the NE5534, which is used in its uncompensated form and is biased into class A using a CCS, performs I/V duties for a TDA1545, the output is fed directly to an all Vishay shunt type stepped attenuator, followed by the amplifier. The power supply for the NE5534 is unregulated and slightly unbalanced though it uses high quality capacitors and shottky diodes. I think it is fair to say that the whole circuit following the DAC IC is very minimalist and it has proved to be very revealing indeed.
Fmak has mentioned the LM6171, I agree that this is a very good sounding op-amp (suitably bypassed) and in my view it is perhaps tonally more accurate than the NE5534 and it certainly has a firmer sound (bass), so firm you could almost stand on it!, though it conveys only a small fraction of the recorded space and ambience that the NE5534 manages, the difference is pronounced. It is this quality that set the NE5534 apart from the others, at least for me. My preference is for real music recorded in a real environment, I don't want the orchestra in my room, in fact I want exactly the opposite, I want to be 'transported' to the venue where the recording took place. So far no FET op-amp has managed to do this for me, of the bipolar types the NE5534 and AD797 stand out from the croud, the former better in some respects, the latter better in others. Unfortunately the AD797 is unstable in this circuit.
I would add that I have achieved a better sound from the NE5534 in the circuit mentioned above (a modified Philips CD723) than I have obtained from a Marantz CD17KI.
Carlos,
You have written the NEs sound ''crap, lifeless, undynamic, spitty, sibillant. I agree with you with regard to ''undynamic'', however I simply cannot agree in any other respect, in fact I would redirect your first, fourth and fifth criticisms towards the OPA627! In my experience the (uncompensated)NE5534 produces some of the smoothest, warmest and most fluid sounds of any IC op-amp. Different ears and different circuits perhaps?
Tim.
Your reply to my previous post was both refreshingly polite and objective.
The amplifier is a modified JLH class-A, details of which can be found at... http://www.tcaas.btinternet.co.uk/jlhupdate.htm
The speakers are (recently completely rebuilt) Quad ESL-57s, the NE5534, which is used in its uncompensated form and is biased into class A using a CCS, performs I/V duties for a TDA1545, the output is fed directly to an all Vishay shunt type stepped attenuator, followed by the amplifier. The power supply for the NE5534 is unregulated and slightly unbalanced though it uses high quality capacitors and shottky diodes. I think it is fair to say that the whole circuit following the DAC IC is very minimalist and it has proved to be very revealing indeed.
Fmak has mentioned the LM6171, I agree that this is a very good sounding op-amp (suitably bypassed) and in my view it is perhaps tonally more accurate than the NE5534 and it certainly has a firmer sound (bass), so firm you could almost stand on it!, though it conveys only a small fraction of the recorded space and ambience that the NE5534 manages, the difference is pronounced. It is this quality that set the NE5534 apart from the others, at least for me. My preference is for real music recorded in a real environment, I don't want the orchestra in my room, in fact I want exactly the opposite, I want to be 'transported' to the venue where the recording took place. So far no FET op-amp has managed to do this for me, of the bipolar types the NE5534 and AD797 stand out from the croud, the former better in some respects, the latter better in others. Unfortunately the AD797 is unstable in this circuit.
I would add that I have achieved a better sound from the NE5534 in the circuit mentioned above (a modified Philips CD723) than I have obtained from a Marantz CD17KI.
Carlos,
You have written the NEs sound ''crap, lifeless, undynamic, spitty, sibillant. I agree with you with regard to ''undynamic'', however I simply cannot agree in any other respect, in fact I would redirect your first, fourth and fifth criticisms towards the OPA627! In my experience the (uncompensated)NE5534 produces some of the smoothest, warmest and most fluid sounds of any IC op-amp. Different ears and different circuits perhaps?
Tim.
The speakers are (recently completely rebuilt) Quad ESL-57s, the NE5534, which is used in its uncompensated form and is biased into class A using a CCS, performs I/V duties for a TDA1545
-------------------------------------------------------------------------
Seems to me too slow to perform I/V . You have, of course, a classic system.
-------------------------------------------------------------------------
Seems to me too slow to perform I/V . You have, of course, a classic system.
I'm pretty sure you would all freak out if you heard my four channel bi-amping gainclones... which use LM12s! Haha. With these bad boys it doesn't really matter if you use 5532s or whatever. Actually it sounds surprisingly good though! For my more serious equipment I'm still undecided about the 5532 or the OPA627, so this thread has been great for me so far. Keep the opinions coming (both subjective and otherwise). I guess all opinions are subjective anyway.
Pete
Pete
carlosfm said:Tim, I really can't figure out how can you put an OPA627 sounding bad?
Poor PSU bypass?
Of course.
The NEs will deliver in any crappy circuit.
Put a good (and fast) op-amp and it's disaster.
Take care of a proper bypassing an you'll see the light.
we found just 1 very HQ electro from each PS pin to gnd plane
was best with NO parallel ceramic or film bypass cap. These form a
tank circuit and increase HF edge. This was fine for 627 or
637 however with faster devices probably just an electro
is insufficient. With pure battery supplies and no active
regulation we found the 627/637 with 10mA class A OP bias
delivered VG sound.
can't really comment any more on 5534 till I re try it. will
have a new dac running in a week or two so that should be
a nice test bed.
cheers
terry
Carlos,
You wrote ''I really can't figure out how can you put an OPA627 sounding bad?''
I think this can be attributed to me general dislike for the sound of FETs, though the OPA627 has an impressively detailed sound.
I see your point with regard to bypassing but the OPA627s were bypassed with low impedance caps in the supply and 100nF across the supply pins.
A question for you Carlos...
In what circuit do you use the OPA627?
Have you tried an AD797, if so what do you think of it?
Terry,
Can I ask you the same question wrt the AD797 please?
Tim.
You wrote ''I really can't figure out how can you put an OPA627 sounding bad?''
I think this can be attributed to me general dislike for the sound of FETs, though the OPA627 has an impressively detailed sound.
I see your point with regard to bypassing but the OPA627s were bypassed with low impedance caps in the supply and 100nF across the supply pins.
A question for you Carlos...
In what circuit do you use the OPA627?
Have you tried an AD797, if so what do you think of it?
Terry,
Can I ask you the same question wrt the AD797 please?
Tim.
fmak said:
No, it's not black art.
I always do it this way:
22~100uf electrolythic + 0.1uf ceramic (or polyester) from + to ground and - to groud, NEAR the op-amp's PSU pins.
0.1~1uf from + to - UNDER the op-amp.
High-End.
Now put the NEs back and you won't believe how crappy it sounds against an OPA604/2604, OPA132/2132, OPA228/2228
, OPA627, LM6171/2.Good bypassing applies and is good to ANY op-amp.
If you have good bypassing and the NE sounds like crap, then it IS CRAP .
TimA said:
Hi Tim,
The 797 for I-V sounded good, sort of very linear and smooth,
but flat, it wasn't nearly as good as 627.
I also tried 797 in a high gain mic pre application and didn't
like it much. Went completely discrete.... game over, bye
bye opamps. I've donr this a few times now with mic pre's
and don't bother with opa's any more.
Tim, you have to be really careful with power supply
bypassing. Even brand of caps has a profound
effect. Try and stay away from steel leaded low Z types,
they can sound grainy, however I don't mind copper leaded
oscons.
There are other things to consider, the impedances looking
OUT of each OPA IP should be equal. That includes if you have
active filter. This can be done and helps most opa's. FETs often
suffer CM distortion due to IP cap modulation so balancing Z's
at IP's will help these.
cheers,
Terry
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.