TPA6139A2: Problem with gain setting resistor

Part Number: TPA6139A2

We are using the TPA6139A2 with a 3.9 kohm (1%) gain setting resistor. I have measured the resistor value to be 3903 ohm.
The gain in the TPA6139A2 should then be 20 dB. However, on our recently produced boards it does not work. The gain seems to be the default (6 dB). When I add 100 ohm resistance (totalling 4.0 kohm) the gain becomes 20 dB. This is however outside the datasheet specification (which states 2 % tolerance on gain resistor value).
We have not seen this behaviour on our 10 previously produced boards.

Component marking for TPA6139 on the old boards: 2BTG4 CHN.
Component marking for TPA6139 on the new boards: 14TG4 C5VT.

As a side note:
In the datasheet it says "Gain setting is latched when the MUTE pin is set high". This does not seem correct, the only time it is latched (as far as we can see) is at power up.
The datasheet also reference "Rfb", which I believe is a copy paste error from some other datasheet.

Regards, Jonas

11 Replies

  • Hello,

    Welcome to E2E, thanks for your interest in our products.
    My colleague will be responding soon with further information.

    Best Regards
    José Luis Figueroa
    Audio Applications Engineer
  • Hi Jonas,

    It seems that this behavior is an isolated event (I haven't seen it before on my tests). Perhaps we could set any failing parts into FA process.
    Regarding the data sheet reference to Rfb and gain setting being latched at MUTE/POWER I will check this out and make any needed change to the document. Thank you for pointing this out!

    Best regards,
    -Ivan Salazar
    Audio Applications Engineer - Low Power Audio & Actuators
  • In reply to Ivan Salazar:

    Hi Ivan and thanks for the quick reply.

    It is very urgent for us to resolve this since customers are waiting for us to ship the product.

    Can I do anything (like measurements) immediately that would help?

    Should i desolder a part and send it to you?


    Regards, Jonas

  • In reply to user4497266:


    Is the gain setting behavior (the one that needs 4kOhm resistor) present on a single device? or are there several devices showing this behavior?
    Have you tried to change the device on the board to see if the failing behavior follows the device?
    I will contact my team to consult the details about the FA. Do you have an email address I can use to message you directly?

    I've verified that gain latch is performed at power up and not when disabling the mute. I will work on getting this corrected as well as Rfb reference.

    Best regards,
    -Ivan Salazar
    Audio Applications Engineer - Low Power Audio & Actuators
  • In reply to Ivan Salazar:

    It is present on several (at least 10 I would say) devices.

    Each of our boards have 4 devices and on one of them, 2 of the devices work but the other 2 don't.

    Since it is production boards I have chosen not to modify them too much. But I will change 4 devices on one of the boards tomorrow morning and see what happens.

    You can mail me at:

    Regards, Jonas Aberg

  • In reply to user4497266:

    We have now changed 4 devices on a board and all changed devices function properly.
    Do you want me to send you the 4 desoldered devices?
  • In reply to Ivan Salazar:

    Hi again.

    Do you have any update? I have not received any mail from you yet.  Regards, Jonas

    [Handled off-line  Issue resolved]

  • In reply to user4497266:

    This issue is not resolved by any means!

    Trying different resistor values until it seems to work is not an acceptable answer.

  • In reply to user4497266:

    This will need to be handled through Failure Analysis.
    Could you please contact your distributor for returning the parts?

    Best regards,
    -Ivan Salazar
    Audio Applications Engineer - Low Power Audio & Actuators
  • In reply to Ivan Salazar:

    The parts are mounted on PCB's. As I mentioned earlier, I have desoldered 4 parts that I can send to you (or via our distributor).

    But we have another batch of TPA6139 from another distributor that also does not work as described in your datasheet. Should we return these instead?