Mark C wrote:Thanks Jim for the information.
I looked at that diagram before I posted the question. Now after looking at it again, I am totally unable to figure out how you got 16 or 32 ohms.
According to the datasheet, the TLV320 output Z at the headphone output pins is close to 0 ohms -- like many active amp stages. The spec sheet shows
"Headphone-amplifier output load resistance" of 0 as a minimum but that doesn't mean you operate it that way, nor does it represent the output Z at the headphone jack. The reference to 16 and 32 ohms on the datasheet is the maximum power delivered into those resistances.
As an example, the Orion MKII board (and I believe Angelia) has 10 uF caps in series with a tapped RF choke between the TVA320 and the audio jack. Those components form part of the output Z at the headphone jack and it varies with frequency but it's always more than zero. Not having the G2 schematic set, it's only a guess on my part, but I suspect Apache kept the same headphone output circuit.
As an example of the Orion board at 100 Hz, the reactance at the headphone jack is -160 ohms in series with the choke resistance, let's say 10 ohms. The Z is 10-j160 but since the reactive component is much more than the resistive component, the result is slightly more than 160 ohms at 100Hz. But at 4 kHz, it drops to 10-j4. Now, the resistive component exceeds the capacitive component for a total output Z of about 11 ohms.
Bottom line: any modern headphones have sufficient impedance. The output jack Z changes with frequency just as headphone impedance changes with frequency. My Sony MDR-7506 headphones have an impedance spec of 24 ohms but Sony correctly caveats the impedance at 1 kHz although it's much different at the audio spectrum extremes.
Paul, W9AC
P.S.
The RF choke has a small amount of inductive reactance that cancels the much larger cap reactance, but it's small enough to neglect at audio frequencies.