Audio Buffer – unity gain link stage/headphone amp with ultimate output drive capability

[March Audio]

What does it look like into say a 10 k load plus a small capacitance?

 

[pma]

What does it look like into say a 10 k load plus a small capacitance?

Virtually same as into 1Mohm.

I placed the PCB to the Modushop box to make a headphone amplifier.

Edit: 115kHz square response + rise/fall time added

 

[pma]

I made a measurement with RightMark SW, a comparison between Topping D10s-E1DA Cosmos loopback and the same with the Audio Buffer inside the loop.

Frequency response

Noise level

Dynamic range

THD

IMD

Crosstalk

We can see that the Audio Buffer makes no audible degradation. If we explore THD spectrum of individual measurements, then with Audio Buffer inside the loop we have cleaner spectrum with much less high order harmonic distortion components. This because the output stage of the Audio Buffer is able to give much more output current and operates at higher idle current than the output opamp of the D10s DAC.

 

[restorer-john]

The in band ripple is more than I would expect. I presume it is the D10s.

Nice buffer Pavel.

 

[pma]

The in band ripple is more than I would expect. I presume it is the D10s.

Nice buffer Pavel.

Thank you .
The in band ripple is mostly from D10s, but also from E1DA. We pay for the very steep filters in both devices, the price of almost no mirror images.

Even more ripple for 44.1kHz Fs.


Buffer open photo (as tested)

 

[El_Arte]

@pma Pavel, are you going to sell these boards or share the Eagle/Gerber files?

 

[pma]

@pma Pavel, are you going to sell these boards or share the Eagle/Gerber files?

I was selling PCB boards in the past, but I stopped about ten years ago. I am sorry I do not share Gerber files, the only exception is Openamp - phono preamplifier.

Preamplifier for MM cartridge

 

[El_Arte]

OK

 

[pma]

I have a new analog generator, described in this thread. I was curious how audiobuffer would measure at 7V input level. I first tested it with several op-amp: OPA134, OPA627 and AD825. AD825 added considerable distortion, OPA134 added some distortion and only OPA627 was able to follow thee generator parameters. In fact, it has been an improvement to the generator capabilities. Further measurements are made with the OPA627 as an input op-amp of the audiobuffer.

1. Generator spectrum measured directly with Cosmos ADC


2. Generator + audiobuffer

Please note reduced H2 and improvement in THD+N, as a result of the fact that the generator is now loaded with high input impedance of the audiobuffer, instead of the low input impedance of the Cosmos ADC.

3. Generator + audiobuffer + 50ohm load resistor

THD is only slightly higher then without the 50ohm load. THD+N rise (and degraded SNR) is now due lower output voltage (50 ohm audiobuffer output impedance + 50R load) and constant Cosmos input noise.

4. Generator + 50 ohm load directly

This does not work. The generator with 5532 devices cannot drive 50 ohm load.

5. Conclusion
The circuit like the audiobuffer improves generator driving capability without worsening generator parameters. It would have been nice if Cosmos had an option of high impedance input.