2
\$\begingroup\$

I managed to reverse engineer a headphone amplifier in a USB DAC that I bought on aliexpress. As I expected there is a electrolytic capacitor in series to the output, but there is also a 10Ω resistor in series. What is purpose of this resistor? I would like to short it to increase damping factor (see NOTE1), but I'm sure there is a reason to put it there.

DAC is "called": SPDIF Interface USB decoder ES9028Q2M + AD823 + SA9023 ES9038 DAC computer Sound Card headphone amplifier not need power supply

NOTE1: Headphones have unwanted own resonances. Removing the series resistor decreases their effect to the sound assuming the headphones are passive.

enter image description here enter image description here

\$\endgroup\$
5
  • 1
    \$\begingroup\$ Quite possibly helps short circuit protection (for the amp). Headphone jacks are notorious for shorting things while plugging in... \$\endgroup\$ Sep 14 '20 at 15:39
  • \$\begingroup\$ Did you check the datasheet for the amplifier? Very often the circuits are copied directly. \$\endgroup\$
    – Transistor
    Sep 14 '20 at 15:39
  • \$\begingroup\$ You haven't indicated what power rails the op-amp runs from. \$\endgroup\$
    – Andy aka
    Sep 14 '20 at 15:44
  • 1
    \$\begingroup\$ What's "dumping factor"? \$\endgroup\$
    – Andy aka
    Sep 14 '20 at 15:49
  • \$\begingroup\$ I thought about short circuit protection, but AD823 already has it. I don't have datasheet for this amplifier, in Linux it's detected as SA9023, which I belive is name of the USB controller. Op-amp is powered from USB through choke. \$\endgroup\$
    – urhen
    Sep 14 '20 at 15:57
2
\$\begingroup\$

It is there for many reasons, but we can't really know what was in the head of the designer when this was drawn.

  1. The output is unstable with capacitive load. The resistor isolates the output capacitance from op-amp.

  2. Short circuit protection. It protects both the op-amp and headphones.

  3. Crude volume limiter.

  4. Op-amp output does not like to be driven into saturation. Saturation load is 25 ohms according to the datasheet. With 10 ohm in series, it can drive 16 ohm headphones without problems.

\$\endgroup\$
1
  • \$\begingroup\$ Since Figure 26 in the AD823 datasheet shows how to choose the value of that resistor to stabilize the op amp when driving capacitive loads, I think #1 is the most likely explanation. For lower capacitance headphones, that resistor can likely be removed. \$\endgroup\$ Sep 14 '20 at 16:40
1
\$\begingroup\$

It's likely that it is being used to limit some of the high frequency peaks. I am not an audio expert and don't know if that is generally a common thing for audio performance.

By limiting the output current we can limit the ramp rate of the signal - this usually means that the output waveform is more rounded. When looking at the waveform in the frequency domain the harmonics are attenuated.

After reviewing the datasheet it looks like it is recommended to add some series resistance to increase phase margin of the opamp.

If greater phase margin is desired, a small resistor can be used in series with the output to decouple the effect of the load capacitance from the op amp (see Figure 26). In addition, running the part at higher gains also improves the capacitive load drive capability of the op amp

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

It's a protective resistor to limit maximum power delivered to the load so you're less likely to blow up your headphones or your eardrums.

\$\endgroup\$
3
  • 2
    \$\begingroup\$ 10Ω resistor with 60Ω headphones lower volume only by about 1.34dB. I connected oscilloscope to the output of the amplifier and it showed about 10mVrms, since it has power supply of 5V, it has plenty of headroom to damage something. So I don't think this is the right answer. \$\endgroup\$
    – urhen
    Sep 14 '20 at 16:06
  • \$\begingroup\$ Typical low cost consumer headphones have impedances more like 15-30 Ohms (roughly). \$\endgroup\$
    – mkeith
    Sep 14 '20 at 18:34
  • \$\begingroup\$ If you're using 16 Ohm earbuds and cranking the level - it looks like max levels can exceed 1V - you will be thankful for that 10 Ohm ear protector. \$\endgroup\$
    – td127
    Sep 14 '20 at 18:38
0
\$\begingroup\$

Headphone cables can have a lot of capacitance (as much as 0.5nF+) and that is a pretty high bandwidth amplifier, and the gain is close to worst-case.

So the 10\$\Omega\$ provides a fair bit of insurance, though I suspect it will typically be pretty good with that relatively huge 100pF feedback capacitor in there, and simulation seems to confirm that. With a very small cap rather than 100pF it has a tendency to ring at 5 or 6MHz.

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

This is a guess, I haven't measured nor calculated it.

AD823 is high speed amplifier, it can amplify at several megahertzes. That's limited by inserting those small capacitors, but together with complex reactive load (speaker element coil inductance + wire capacitance) the total effect can be difficult to predict. The series resistance guarantees the amp is never loaded by a low loss resonant circuit.

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

The main reason may be to allow the output to work reasonably well both with headphones and with line level inputs. Line level inputs usually have high impedance and so will receive high levels. Headphones typically have much lower impedance and so will have some gain down effect from the 10 Ohm resistor in series.

The other answers are good, and may be more correct than this one. But nobody else mentioned this so I though I would.

\$\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.