Direct coupling without servo/global feedback?

[EUVL]

The best buffer with no unobtaniums we have (unity gain) :
The XEN XELF Headphone Buffer

It is thermally stable because of the lateral FETs' negative tempco.
It is also very easy to drive.
The signal source only sees the gate of 2 lateral FETs and 2 current sources in parallel.
And it will easily drive 8R speakers.


Patrick

 

[lasercut]

Thank you for all the suggestions, the XELF looks perfect.

 

[sciencesky]

This design by Wayne Kirkwood is direct coupled (all DC) and only local feedback within the THAT1646 or DRV134 driver IC. Overall, no global feedback. It sounds and measures excellent. One of the lowest distortion headphone amps I have measured.

Simple Class A Headphone Amp Using THAT1646

I like that concept. I think I have an idea using a THAT1580 in headphone amp.
It wouldn't be ZFB but it would be something interesting to try.

 

[EUVL]

There are much simpler ways for biasing the output devices.
Just move the feedback loop to pin2, and you get the same thing.
The THAT1646 only does this for you inside the chip.

For headphones and normal opamps, just change the 4.7R to 470R.


Cheers,
Patrick

.

 

[sciencesky]

Yes you can find that concept also on Elliots website but also with diodes biasing the transistors.

 

[xrk971]

Here is another DC coupled design with all discrete components. It was designed by member TimS. It does have a feedback path, but not a traditional topology. It sounds great.
Simple High Performing Headphone Amp

It was a perfect candidate for a home-etched PCB.

I tweaked some of the resistor setpoints to allow use of all BF862’s. With adjustments, it would work with all 2SK209’s as well vs mix of J111 and 2SK209. A very transparent highly resolving, and dynamic amp. It’s Class A and has a dominant second harmonic output but overall low THD.

Using a DCDC based PSU cleaned up some of that PSU mains noise