0
\$\begingroup\$

I want to use a "capacitor-less" headphone amplifier that doesn't require DC blocking capacitors at the output.

I found a headphone amplifier that has a negative charge pump that generates an internal negative supply voltage, which removes the need for output caps. However, I found these parts are more expensive. As an alternative, I found a lower cost headphone amp that has a "phantom ground" configuration to eliminate the need for output coupling caps, as shown below:

enter image description here

Is there any potential issue with directly connecting the phantom ground (Vout3 in the diagram) to the headphone's ground? What's the disadvantage of this method over the negative charge pump method?

\$\endgroup\$
3
  • \$\begingroup\$ The headphone out jack should be an isolated jack, because the Vout3 is the headphone output "ground" which in reality is an inverted summed signal. you would never connect the Vout3 to the headphone's input "ground". The reason why I express ground like "ground" because its slang for the common. because AC has common , not ground.I don't know where they get phantom ground, because powering something by phantom power is injecting DC on a AC signal line to power electronics somewhere else. \$\endgroup\$
    – drtechno
    May 29 '19 at 4:42
  • \$\begingroup\$ I'm confused. This is the datasheet for the part (TS4909): st.com/content/ccc/resource/technical/document/datasheet/f0/25/… It can be operated with or without phantom ground (there's a selection pin for that). \$\endgroup\$
    – donut
    May 29 '19 at 4:57
  • \$\begingroup\$ oh I see they brought out the non-inverting pin. well, I would test both methods and see what is the lowest noise configuration with the power supply \$\endgroup\$
    – drtechno
    May 29 '19 at 14:00
1
\$\begingroup\$

I believe the correct term here is the virtual ground. "Phantom ground" sounds like a mix of "phantom power" and "virtual ground".

Virtual ground does have downsides, see for instance Virtual Grounds & 3 Channel Amps. It's fine for low current applications, which may include a headphone amplifier, depending on its peak output power and the virtual ground implementation. If the implementation is not good enough, it may cause crosstalk between channels.

Whether it is better or worse than a switching power supply really depends on concrete implementations of both.

\$\endgroup\$
1
\$\begingroup\$

The IC is designed to use the "phantom ground" so use it. A charge pump adds noise.

\$\endgroup\$
0
\$\begingroup\$

This provides a lower impedance at Vcc/2 than is possible with a charge pump since it is DC coupled while the charge pump is created by AC cap coupled diode clamps and doublers.

There is no problem with connecting a headphone and I suspect it has a floating DC power supply so the DC midpoint is floating with respect to earth ground and would also have no problems connected to an earth-grounded Line input as well.

The only difference with other DC coupled amps is it has 3 amplifiers instead of 4 in bridge mode so the output voltage is single-ended vs differential output as in "bridge-mode" amps which are also DC coupled. Bridge amps are advantageous into 2 Ohm sub-woofers with lower Vcc.

But being that headphones don't need much power, so it's not a problem. It's a solution!

It is best to compare all specs for THD, power, load impedance, BW and compare cost vs performance. But in theory, should perform better at a lower frequency and lower impedance headsets.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.