I am baffled. I think I miss red the info on the Burson OpAmp regarding the technicalities. I will re check asap and let you know
I've downloaded the datasheets of both the
NJM4556A (NJM) and the Burson Audio V5i (BA, see posting #8) and the Burson is no match:
- The NJM can deliver up to 70 mA output current into a 150 Ohm load at 10.5 V. There is no maximum output current specified at all for the BA. The only specification for a load is 600 Ohms, 4 times higher than NJM and at least 2 times higher than common high impedance headphones.
- Looking at the mechanical construction of the BA I would never use it for driving headphones since the shield prevents dissipation of heat of the SMD chip underneath (no exchange of air required for cooling). Since BA also does not specify whether it has thermal protection it is possible that listening at higher SPL may destroy it.
Agreed, NJM does not specify this either but you can rely on Sabaj to have taken this in account. Otherwise they would get a lot of complaints and warranty repairs.
- The NJM has THD of below 0.0003% at ~4 V output (the sweet spot, at lower voltages noise dominates) to a load of 200 Ohm, and this at 30 dB gain (which means you can expect lower THD for lower gain). The BA specified 0.005% (~ 15 times higher!) at a lower voltage and a lower load, and the gain is not specified so you should expect any lower THD than this.
- The NJM specifies a voltage noise of less than 10nV/sqrt(Hz) between 20 Hz and 1 kHz. The BA does not specify noise at all which usually is a hint that this spec is bad (yes, you have to read between the lines, even in data sheets of respected manufacturers). The existance of FETs in the input stage is another hint.
- Even BA does not list the NJM4556 as a possible candidate for replacement.
Conclusion: Don't do it, you won't improve the sound, especially not with the Burson.
Nonetheless, I wanted to try the Burson OpAmp, as I am curios how it could sound, ...
You can't. To compare its sound you need a second unmodified D4 to be able to perform a reliable double blind test (DBT). Otherwise you may hear a lot of differences due to expectation bias and the inability of us humans to remember sound. It's the way we are wired, all of us. See my failures below.
...specially with a great device such as the Sabaj D4.
The Sabaj D4 may be great because its designers used the best matching components they could think of. Why do you think anybody else would be able to do better than them without insight into their development documentation or their minds?
I know what I'm talking about. Although I am an EE I fell myself victim to op-amp rolling in an Arcam Black Box 3 DAC and thought it sounded much better than before, until I did a blind test against an unmodified model where I failed miserably. Recent measurements opf this DAC showed that the performance of the modded model was not better although I used an op-amp which was very much hyped by audiophiles and the press at that time. You can read about it
here and
here.