[Spacehead]

Quote:

Originally Posted by ericj /img/forum/go_quote.gif I still don't understand what you're on about wrt the feedback resistor on the ground channel.

TLE2426 output=virtual ground

virtual ground -> op amp -> buffer -> output ground

this time I ditched the three op amps and used sijosae's class ab "buf634" buffer instead. It needs to be kept in op amp control, in feedback loop or there will be dc offset.

But now I understood, I don't need a resistor, I need only bare wire, right? hmm, I think I should start practicing using some schematic and simulation program (know any good and easy for Windows?), because I sometimes do things more complicated than is necessary.

 

[ericj]

Quote:

Originally Posted by Spacehead /img/forum/go_quote.gif TLE2426 output=virtual ground virtual ground -> op amp -> buffer -> output ground this time I ditched the three op amps and used sijosae's class ab "buf634" buffer instead. It needs to be kept in op amp control, in feedback loop or there will be dc offset. But now I understood, I don't need a resistor, I need only bare wire, right? hmm, I think I should start practicing using some schematic and simulation program (know any good and easy for Windows?), because I sometimes do things more complicated than is necessary.

I suggest examining the schematic of any amp already designed to have a buffered ground channel - like a pimeta, or an M3. Or even a mini3, though that doesn't have a buffer.

 

[diditmyself]

Quote:

Originally Posted by Spacehead /img/forum/go_quote.gif TLE2426 output=virtual ground virtual ground -> op amp -> buffer -> output ground this time I ditched the three op amps and used sijosae's class ab "buf634" buffer instead. It needs to be kept in op amp control, in feedback loop or there will be dc offset. But now I understood, I don't need a resistor, I need only bare wire, right? hmm, I think I should start practicing using some schematic and simulation program (know any good and easy for Windows?), because I sometimes do things more complicated than is necessary.

The problem with BUF634 as ground channel buffer is not DC-offset. It's the high output impedance = bad crosstalk = unnatural soundstage.

For simulation most of us use LTSpice. Since there are many users it's easier to get support from fellow DIY'ers. It's free, and the simulations correlate well with reality.

 

[amb]

Quote:

Originally Posted by ericj /img/forum/go_quote.gif I suggest examining the schematic of any amp already designed to have a buffered ground channel - like a pimeta, or an M3. Or even a mini3, though that doesn't have a buffer.

Or rather -- the Mini³'s opamps have the "buffer" built-in.

 

[Spacehead]

Quote:

Originally Posted by ericj /img/forum/go_quote.gif I suggest examining the schematic of any amp already designed to have a buffered ground channel - like a pimeta, or an M3. Or even a mini3, though that doesn't have a buffer.

I see that my ground channel is almost like PPA ground channel except my JFET cascode is in front of buffer op amps. Shouldn't that pull all of them to class A? I don't know how to calculate the bias , I am using 2SK246 JFETs and 1K resistor.

 

[Spacehead]

Is it possible to further bias this amplifier into class a? I added 78L05 regulators to outputs configured as current source with 3K resistors. Seems to be working.

here is the link: http://www.headphoneamp.co.kr/bbs/zb...desc=asc&no=58


edit: now with 1K resistors, even smoother sound, no distortions anymore at all. I see that in original config 2K resistors are the current sources. But they set transistor bias, not op amp.

Could this be modified to Class A? I wonder will those BC337/BC327 work in class a too long, if they heat up too much.

 

[buzz2086]

I know this is a really old thread. But I have some questions on this headphone amp. There are few versions of BC337 and BC327. They have BC337-25 and BC337-40. Which one is suitable for this amp ?