Buf-03

planet10 said:


Thanx...

i may have to ask about the idled at 15 mA but i should go grab some data and see if i can figure that out myself.

The application is
Aunt Corey's Buffered Pre

dave

Hi dave, sorry for late reply!?

You cannot compare BUF03/04 with BUF634. The Analog Devices parts are currentfeedback design and has very low distortion and wide bandwidth.

Why dont you try built the discrete version of the BUF634? The BUF634 is nothing more than a copy of the W.G. Jung buffer (Diamond buffer).

It has no feedback!

Go for the Walt G. Jung.

Use transistor like 2SC2240/SA970 for buffer and driver and Use 2SA1209/SC2911 as output idling at 20mA.

I would say it will have better sonic performance than the BUF634 alone.
 
By the way i have just stumpled over this link today.

It is a design done with a BUF634 and a AD8620AR with class A bias of the opamp.

and this is the benchmark.

They are not that fantastic like the one done by PMA, but the schematic is not the same either They have done a little change in the feedback loop. They have lowered the overall openloop gain of the circuit, there by revealing more of the BUF634 performance.

Take a look and jugde for you self.

The design done by PMA is indeed good to.
 
Sonny,

regarding the measurement, they evidently did not solve background problem. This a a time consuming work and I do not wonder they probably do not want to solve it. Fortunatelly for me this is done by a colleague of mine (thank you, Vladimir ;) ). There is apparently mains frequency interference and not clear background in their measurement.

Pavel
 

Bricolo

Member
2002-11-11 6:52 am
sonnya said:
By the way i have just stumpled over this link today.

It is a design done with a BUF634 and a AD8620AR with class A bias of the opamp.

and this is the benchmark.

They are not that fantastic like the one done by PMA, but the schematic is not the same either They have done a little change in the feedback loop. They have lowered the overall openloop gain of the circuit, there by revealing more of the BUF634 performance.

Take a look and jugde for you self.

The design done by PMA is indeed good to.

What do they use a rail splitter, instead of simply putting the ground between the 2 9V cells?
 
sonnya said:
By the way i have just stumpled over this link today.

It is a design done with a BUF634 and a AD8620AR with class A bias of the opamp.

and this is the benchmark.

They are not that fantastic like the one done by PMA, but the schematic is not the same either They have done a little change in the feedback loop. They have lowered the overall openloop gain of the circuit, there by revealing more of the BUF634 performance.

Take a look and jugde for you self.

The design done by PMA is indeed good to.

If the BUF634 is Jung's diamond buffer in disguise, then the MINT is almost completely "Junguian".

The MINT arrangement uses two gain chains: one on first stage and a global one. Jung used a third one for his latest chip preamp on the second chip, which was a not a buffer chip as this one is. But the Jung spirit to minimize input errors seems to be there, which I think is great. Walt Jung has regularly helped the audio DIY community with very good sounding projects, and I sure he is happy to see more people implementing them.

In fact the MINT seems like a cross between two Jung designs: the DAC he suggested some time ago and his low input error preamp.

Quite interesting your comments on supply glitches present on the noise graph. Now this is supposed to be a battery design, so how did they get there?

About why they are using a rail splitter when they are putting two batteries, it's a mystery to too. Would a "regular" V+/GND/V- supply improve the specs?



Carlos
 
Bricolo said:
Where can we see Jung's schematics you're talking about?

The original minimum error preamp article can be seen here:

http://www.elecdesign.com/Globals/PlanetEE/Content/1518.html

The other preamp, which was one of the first implementations for a DAC on CD players which was then much imitated, was published in The Audio Amateur a long time ago. It used an NE530 on first stage and an LH0002 on second stage. Jung suggested using that "diamond" discrete arrangement instead of the 0002 on second stage.


Carlos
 
carlmart said:


The original minimum error preamp article can be seen here:

http://www.elecdesign.com/Globals/PlanetEE/Content/1518.html
Carlos

Interesting reading. Thanks!

There is one detail in the paper I don't find quite clear, but which
I think I understand correctly. However, I would appreciate if
somebody could confirm this. Jung says resistor RC should be
choosen s.t. RC = RS. The only way I can understand this is
that it is not the exact answer, but a practical approximation
which is sensible if (R1||R2) << RC << RD, which can be
considered to hold with the actual values given in the
diagram. (I think the exact answer would be something
like ((R1||R2) + RC) || RD = RS ).