Which audio buffer design is the best for Gainclone?

Status
Not open for further replies.
Jump to Latest
#101
Nuuk said:
Oscon SG capacitors are equivalent to much higher values in other capacitors so those 15 uF Oscons are more like 150 uF. However, I will try some Panasonic FC's in yet another new circuit!
Click to expand...

OsCons are good for digital (Dacs, etc.).
I think the Panasonics may be better here. Most of the time I use 47uf/25v or Philips 68uf/25V, wich I can find easily and work very well.

Nuuk said:
I note that some people put the electrolytic decoupling caps nearer the regulators and have smaller film caps (0.1 uF) right on the pins of the opamp. Do you recommend that approach?
Click to expand...

No. The OPAs, and one of them is the 627, like capacitance on the PSU pins.
I use 0.1uf on the regs and the electros on the op-amp (bypassed with 0.1uf ceramic, as I always do).

Nuuk said:
I am a little puzzled as I have separate regulators for each opamp circuit. Did you realise that or am I missing something else? :xeye:
Click to expand...

I'm sorry Nuuk, it's that in your page - http://www.decdun.fsnet.co.uk/gainclone2.html#gcbupdate (oh, my name is there😎 ), you have a pic of your buffer circuit and I only saw two regs.
Now I check again and I see that the board is for only one channel/op-amp.:clown:
Forget the resistors then.

Nuuk said:
Thanks for your pointers Carlos. I will give them a try with an open mind although I do feel that the discrete buffer is probably the better solution for this particular job! 😉
Click to expand...

It's possible, why not?
But when everything is right, I have the same oppinion as Pavel:
PMA said:
Only very very well designed discrete circuit is capable to give same sound quality as OPA627 with increased Iq.
Click to expand...
Check his post #14 here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=38984&perpage=10&highlight=&pagenumber=2

Maby you found suck a circuit.
 
#104
Nuuk said:
If you don't have a web site of your own, you will have to be named on mine!
Click to expand...

No time for that now...:bawling:
Work and more work, soldering iron and more soldering iron...😀

Nuuk said:
I am not concerned how clever my buffers are. 🙄 Seriously, what is 'increased Iq"? :xeye:
Click to expand...

Pavel tested the OPA627 biased into class A at around 10ma, with the resistor method, as I said on that thread.
"Increased Iq" is just that, it's not usual to go as far as 10ma...
Clever or not, it just sounds good.😀
 
#105
OPA6x7 and Class-A

I try to resume (and translate from german) the input, I got from a experienced engineer. Sorry about my limited english. Dont hesitate to ask me, if something is not clear.

Here it is:

Most opamps work with class AB powerstages. So, they work in push-pull class A until the output current is bigger than the idle current.

As the datasheet does not give the "inner" values of the OPA6x7, we must estimate the idle current from its powerstage. Let's estimate half from the total idle current (7mA), so 3.5mA.

This means, with an output below +-14V and a load >4k (or <2nF for 20kHz) the chip works in (push pull) class A!

What happens, when you let flow 3.5mA current from the output to Vcc-?

Then the output transistor at the negative side works in class B mode (!?!) and works absolutely not linearly.

When you increase this current (between the output and Vcc-), then you reduce the possible swing to 50%. So the "active" output transistor will work in single end Class A, what produces more disturbance than push pull class A. And the driverstages dont work symetrically too.

Then, dont forget about the thermal feedback.

And the impedances for the halvewaves will become different...
Click to expand...


Do you still prefer this sound? Are you sure, it is not just the additional disturbance, you like?

Maybe, then the better choice would be, to build some tube amps, real class A?

Smile
Franz

Be critical and careful in live!
or
Look and smoke, but look what you smoke!
 
#106
I tried biasing OPA627 with resistors, but didn't like the sound eventually. The initial impression was that it brings improvement: better defind midrange, much more open sound and good presence.

But I also noticed some sort of tension and sort of nervousness in the presentation, like over-equalisation that wasn't completely natural, maybe some grain too, sibilances were somewhat exaggerated as well.

I'm using BP grade and it sounds OK, without that extra biasing.
 
#108
Re: OPA6x7 and Class-A

Franz G said:
Do you still prefer this sound? Are you sure, it is not just the additional disturbance, you like?
Click to expand...

All the above is all very interesting theory but, if you don't mind me asking, have you actually tried biasing an op amp into class A

If not could I suggest that you give it a try...

Every time I have tried it I have found the results very agreeable ( and I hate additional disturbances...😉 )

I think for best results with a resistor it will be necessary to have a very quiet supply. But in my high fidelity universe this is a good idea anyway despite so called good PSRR.

mike
 
#109
All the above is all very interesting theory but, if you don't mind me asking, have you actually tried biasing an op amp into class A
Click to expand...

Yes, I tried it (maybe you find a positive comment from me in this forum).

It was somekind of fascinating at the first moment. But, then I compared (like everything) with my 300B SE (tube) amp or my small EL95 SE amp.

There was a loss of "air" and soundstage, compared to this tube amps. But something "shiny", but not the truth. Undisputed truth.

And I removed the resistor after some days...

Franz
 
#110
Re: OPA6x7 and Class-A

Franz G said:
Do you still prefer this sound? Are you sure, it is not just the additional disturbance, you like?
Click to expand...

if u like it distorted then keep it like that ! people use turn tables / tube amps for the same reason .

if u want the sound to be like in the original recording then u can use any (!) audio op amp and u wont hear any difference in sound if u will use it as described in the datasheets .

ps
your speakers got much greater influence on sound then an op amp , so i think its better to spend more money on accurate speakers then spending your time with different op amps that u think sound different .
 
#111
if u like it distorted then keep it like that ! people use turn tables / tube amps for the same reason .

if u want the sound to be like in the original recording then u can use any (!) audio op amp and u wont hear any difference in sound if u will use it as described in the datasheets .

ps
your speakers got much greater influence on sound then an op amp , so i think its better to spend more money on accurate speakers then spending your time with different op amps that u think sound different .
Click to expand...

I can subscribe, what you write.

In my case, the difference between NE5534AP (Philips) and OPA6x7 is very marginally.

Coral Beta 8 and Fostex FE103E is much more difference!

And another point of view from you is very good: you must make different boards and read every datasheet and adapt the circuit, to compare opamps. Not just change pincompatibel opamps in sockets (isn't this true, Carlos FM?).

Thanks!
Franz
 
#113
Franz G said:
And I removed the resistor after some days...Franz
Click to expand...

I wonder what supplies you were using

most of my experiments were done with

schotkies followed by C L C L C L C

I bit extreme I admit but I continued adding smoothing untill it sounded as good as a battery.

Adding a resistor to anything but the very quietest of supply lines will inject noise into the o/p which is of course bad news.

PS I do not disagree about the superiority of true class A

mike
 
#118
In some cases people need for pleasure listening more distortion and you too, Franz 😉 , what I say quite seriously, looking at your tube input stage for gainclone - what do you think, that make this stuff ? All hifi is little bit about perform magic, 'cos people love it 😎 . Probably many guys here have an distortionmeter or simulator and they should measure this amp ( 627 ) in normal and shifted ( in class A ) region. Do they it ? No, 'cos better is talking about surpraised sound several days 😀 , without any concrete scientific proof and explanation 😎 .
 
#119
Franz G said:
Not just change pincompatibel opamps in sockets (isn't this true, Carlos FM?).
Click to expand...

Yes, implementation can make much of a difference.
The exact place where you bypass, how you do it, grounding arrangements, etc.
And the OPA627 is picky about this.
Just drop it on an existing circuit without careful attention and it will sound bad.

:bulb: LM317/337 makes a good PSU, curiously with the exact same components as I use with the LM338 for the amps. :bulb:
Seriously!

:att'n: I used 78xx/79xx regs on my amps because I have pre-regulation with the LM338.:att'n:
But the point of using them initially was just because I needed a very tight layout, important on the amp.
And this pays too.:angel:
Oh, IMHO the 78xx/79xx sounds good with pre-regulation, others don't.:att'n:
LM317/337 doesn't sound good to me with pre-regulation behind it.:att'n:

As for fullrange speakers... I couldn't live with them.:whazzat:
They would finally end up as 3 ways.:clown:
Shy bass, too "smooth" treble...:bawling:
I like to listen and understand the music in all it's glory.:hbeat:

Oh, and I love the sound of the OPA627 biased to class A.:hbeat:
 
#120
Re: Re: OPA6x7 and Class-A

sss said:


if u like it distorted then keep it like that ! people use turn tables / tube amps for the same reason .
Click to expand...


you hit the target in this hobby!

In my exposure in a lot of forums where debates rages, all talks just boils down to what distortion one wants to listen to (which some audiophiles prefer with tube amps and phono preamps) ?!?

Some people do not want accurate reproduction or too detailed, thus they want distortion introduce as if to mask the audible bad recordings of some materials.

Some people wanted accurate reproduction - but we wouldn't know what the word "accurate" really mean to them if they have not heard the original recording performance.

The DIY spirit is to share what has been discovered to be good implementation as experienced by our peer. Then test it ourselves and listen to it and decide.

The talk about amount of distortion, conventional wisdom (which many of us also understand) maybe considered but not as an absolute as to say it is a bad implementation because calculation suggest so. Try and see for yourself.

For all we know, all this commercial amps selling hundreds of dollars do have lower distortions and complied with conventional wisdom and yet, only as good as paper weight and show pieces.
 
Status
Not open for further replies.
Top Bottom