• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 2998 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM386: LM386 pin 2 capacitor mod to increase output power

John Jones
Intellectual 380 points
Part Number: LM386

Hi, it's the infamous LM386 rearing its head once again and this time someone online has discovered that by capacitively coupling pin 2 to ground via a capacitor rather than directly allows the chip to output a little bit more power before clipping. It's not much additional power, maybe an extra 100mW into 8 ohm but still every little helps with these funny little amps we like to play with.

Is there any other benefits or risk of other problems (oscillation etc) by doing this? I've looked around the internet and have not come across any schematics showing this trick, the only other mention was by someone on this very forum http://e2e.ti.com/support/audio/f/6/t/233812 You'd think a neat trick like this would be common knowledge, unless it's got some hidden catches (which is why I'm posting this question to uncover any).

Here is the internal schematic, the guy on the internet (I can post a link to the video if anyone wants and the rules allow it) said that by just grounding pin 2 it alters the internal bias slightly hence clipping much earlier.

Thanks.

  • 0
    TI__Guru* 84915 points

    Hi, John,

    I think this practice is only useful in applications with high gain. I mean, in applications that result in saturation or clipping. However, the capacitor at the amplifier inputs won't provide a DC path for current flowing (in this case) in the inverted input.

    The input bias current will flow through the coupling capacitor and consequently it will be charged. This will exceed the common-mode voltage rating of the op amp input and the output will drive into limits. So, if you would like to include this capacitor in your design, you would need to include a pull down resistor (of around 1Mohm) between the capacitor and the non-inverted input. This will provide a discharge path for the input bias current.

    I hope this helps you. Please let me know if you have additional questions or comments.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Intellectual 380 points

    Hi, so it would need the 1M resistor placing here?

    Also here is the schematic used by the person who showed me this configuration, I notice he doesn't have any DC path on the non inverting input either where's I've used a 10k potentiometer with a 1uF DC blocking capacitor at the top (so it would still have a resistive DC path to ground between the non inverting input and ground.

    Thanks.

  • 0 in reply to John Jones
    TI__Guru* 84915 points

    Hi, John,

    The first schematic drawing that you attached is almost correct. Sorry, my mistake, I would mean between the inverted input and GND.

    Whichever the input you select to place the capacitor, you would need to add a DC path with a 1Mohm resistor to GND. This path will ensure that the bias current will have a discharge path.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Intellectual 380 points

    Wouldn't the built in 50k pulldown resistor serve this function?

  • 0 in reply to John Jones
    TI__Guru* 84915 points

    Hi, John,

    You are right, I haven't noticed this different structure in this amplifier. This resistor could work for this purpose, but my worry is how the mirror effect will act in this case. Probably when the capacitor is charged, the inverted input will have an additional offset. This could explain why the output power increases a little bit too.

    After these considerations, I don't think it has another risk to take into account.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Intellectual 380 points

    Hi, what is the mirror effect?

    In the test with pin 2 shorted to ground the top of the waveform (1khz sine wave into 8 ohm) clipped much earlier whilst the bottom remained a clean sine wave. Using the capacitor instead it gave some more headroom allowing it be more symmetrical when it came to clipping.

  • 0 in reply to John Jones
    TI__Guru* 84915 points

    Hi, John,

    Thank you for your details about the testing. I have confirmed with my team that this capacitor won't generate any issue on oscillation or others. The engineers confirmed this headroom on the evaluation board.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Intellectual 380 points

    Thanks for testing! Wonder why the original designers never discovered this trick then?