Can anybody explain why the distortion drops in the red curve at higher output levels?
Page 19
https://www.ti.com/lit/ds/symlink/b...=https%3A%2F%2Fwww.ti.com%2Fproduct%2FBUF634A
Page 19
https://www.ti.com/lit/ds/symlink/b...=https%3A%2F%2Fwww.ti.com%2Fproduct%2FBUF634A
Attachments
Looks like some sort of random cancellation of two distortion mechanisms (one crossover and another current limit related, I guess). You can see a dip in the 16 ohm curve, too. Relative phase of the dominant harmonic should flip around this point.
Don't forget this graph is from a combined BUF634 in the loop of an opamp so there are several distortion mechanisms at work.
This is not the distortion of the BUF634 by itself.
Jan
This is not the distortion of the BUF634 by itself.
Jan
I know Jan, but my QRV09 amp has the same behaviour with a TPA6120a stand-alone.
NwAvGuy: QRV09 DIY Headphone Amp
NwAvGuy: QRV09 DIY Headphone Amp
Definitely not an expert on this topic, but just a thought: Aren't CFAs pretty much a buffer at the input, a Wilson current mirror in the middle and a buffer at the output? The output buffer of TPA6120A2 and BUF634 could share similarities in topology and therefore show similar behaviour?!
There is something fishy in these graphs. For the TPA6120, it is visible in the distortion vs. power graphs (fig 5, 6). That contains a square of the output voltage so it may be just the way it is presented. There is also a distortion vs. output voltage graph (fig 4) and that looks normal.
But for the BUF634, it can be seen in the distortion vs. output voltage graph.
Something is not right here.
Jan
But for the BUF634, it can be seen in the distortion vs. output voltage graph.
Something is not right here.
Jan