Cornelis, what sounds good to me may not sound so good to you and vice versa.
I've used the THS4031/4032 as a buffer before and much preferred it over the 4556.
Also, the LME49960, that Eddie refers to above, sounds nice.
No matter what you do, asking an op amp to drive low impedance headphones directly from its output is going to result in nonlinearity. I have prototyped various op amp based headphone amps and have come up with some configurations that work and sound great. Applying large amounts of negative feedback to grossly nonlinear circuits can provide practical circuits, but the distortion spectra will become more complex as the transfer functions of the active devices (in this case the output stage of the op amp) become more nonlinear.
Is this audible? That depends on a whole lot of stuff. Is it measureable? You bet it is.
Traditional distortion specifications (harmonic distortion and intermodulation distortion) are important benchmarks, but they don't address higher harmonic generation. Back in the days of tube amps, there usually wasn't much to talk about past the third harmonic. Global feedback was not universal and when employed it did not have the very large loop gain that is typical of modern designs (especially op amp based designs). But the more stages that you include inside a global feedback loop, the more complex the distortion spectra will be.
The 49600 is a real buffer, and allows the op amp to operate in class A full time. It also thermally isolates the input stage from the heat generated by the output stage. It also introduces much less phase shift than any op amp. And finally it operates in a much more linear fashion than an op amp driving a low impedance load. So there's very good reasons why it sounds so good.
Is this audible? That depends on a whole lot of stuff. Is it measureable? You bet it is.
Traditional distortion specifications (harmonic distortion and intermodulation distortion) are important benchmarks, but they don't address higher harmonic generation. Back in the days of tube amps, there usually wasn't much to talk about past the third harmonic. Global feedback was not universal and when employed it did not have the very large loop gain that is typical of modern designs (especially op amp based designs). But the more stages that you include inside a global feedback loop, the more complex the distortion spectra will be.
The 49600 is a real buffer, and allows the op amp to operate in class A full time. It also thermally isolates the input stage from the heat generated by the output stage. It also introduces much less phase shift than any op amp. And finally it operates in a much more linear fashion than an op amp driving a low impedance load. So there's very good reasons why it sounds so good.
Last edited:
After reading a lot about headphone amps I decided to take the plunge and build a cmoy. But it's hard to find a kit in India except on ebay.
I found one site selling a kit,
CMoy Head Phone amplifier kit Is this any good?
The site seems new but I can see a few reviews.
Experts please advise.
I found one site selling a kit,
CMoy Head Phone amplifier kit Is this any good?
The site seems new but I can see a few reviews.
Experts please advise.
@skd123 I've built CMOYs using TL072
,4558, & also more acceptable by DIYers "The OPA2134". If you read datasheets, most opamps are spec. at 600R or above loads. JRC4556 is always the winner. I used scope to observe clipping behaviour @33R load.
Hint- Use +/- 5V to power CMOY with gain not exceeding 3X.
Sent you PM.
,4558, & also more acceptable by DIYers "The OPA2134". If you read datasheets, most opamps are spec. at 600R or above loads. JRC4556 is always the winner. I used scope to observe clipping behaviour @33R load. Hint- Use +/- 5V to power CMOY with gain not exceeding 3X.
Sent you PM.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.