TS3A227E: TS3A227E: negative pulse cause pop with line out to power amplifier

Part Number: TS3A227E

Hi team:
Issue:
When plug a line out line to amplifier, the negative pulse will cause a big pop. The pulse can be observed even just plug a 3.5mm jack only (without lines).

Schematic:

Register settings:
p_ts3a227e->TS3A227E_I2C_WRITE_FUNC(0x04, 0x80);
osDelay(100);
p_ts3a227e->TS3A227E_I2C_WRITE_FUNC(0x03, 0x00);
p_ts3a227e->TS3A227E_I2C_WRITE_FUNC(0x04, 0x26);
p_ts3a227e->TS3A227E_I2C_WRITE_FUNC(0x05, 0x04);
p_ts3a227e->TS3A227E_I2C_WRITE_FUNC(0x06, 0x39);
Thanks a lot for support,
Kevin Dang

8 Replies

  • Kevin,

    Thank you for the screen shot and diagram. Is it possible to send a better resolution image of the schematic I'm having difficulty reading the words. Is it possible for you to label where you are taking the scope shot and where you are connecting to the amplifier?

    This also maybe an issue with translation but can you label or explain where the "lines" are and tell us what it means to have the lines or not have the lines.

    Thank you,
    Adam
  • In reply to Adam Torma:

    Adam,

    Sorry now it's the schematic we can get, I can't find out issues with the schematic, it works fine with kinds of headphones. Use headphone/headset the pulse can't be heard. The measured pulse is Socket's GND to real GND. the L/R output has no pulse related to real gnd.
    Could you please help to check this on EVM, the business is >300ku.

    Thanks a lot,
    Kevin Dang.
  • In reply to Kevin Dang:

    Adam,

    Customer get same result with TS3A227E evm, test data attached,

    Please help that the product is near MP, few days left,

    Thanks a lot for your support,

    Kevin Dang

    TS3A227E+pop+noise+issue.pdf

  • In reply to Kevin Dang:

    Kevin,

    Thank you for the additional information. 

    What it looks like you are observing is the TS3A227E automatic detection sequence.  When an audio accessory is inserted into a device the TS3A227E will automatically conduct a detection sequence to determine if the audio accessories is a 3 pole or 4 pole device and where the ground and mic paths are located. 

    The TS3A227E detection sequence will inject a small current (~100uA) from the tip pin to ring2 and tip pin to sleeve to compare the resistances between Ring2 and Sleeve.

     

    This link e2e post will help explain the detection sequence in more detail . 

    Does TS3A226AE support high impendance microphone (like IPhone 6 earphone ) ? (this post also explains the how the TS3A226AE audio jack detection works

    As well as this app note TS3A227E and TS3A225E Selfie Stick and Accessory Detection Overview

    e2e.ti.com/.../1959790 iphone#1959790

    To help confirm this is what you are seeing you can disable the automatic detection sequence in register 04xh.  Disabling the automatic detection should remove these two ramps.

    I don't recall that the small 100uA current injected onto the tip, ring2, and sleeve pins were at 1.75 V and audible.  Do you know what the impedance of the speaker is or if the signal could be getting amplified inside the customer device? 

    Thank you,

    Adam

  • In reply to Adam Torma:

    Adam,


    Generally the amplifier has 2k~10k input impedance and generally voltage gain > 10v/v.

    Do we have some workaround to minimize the impact.

    Thanks a lot,

    Kevin Dang

  • In reply to Kevin Dang:

    Hello Kevin,

    As Adam posted in the previous thread, the observed behavior you are seeing in the pdf you posted seems related to the detection sequence of the TS3A227E.  With respect to the TS3A227E, the TS3A225E is quite similar in design and its detection is done in two steps outlined below (figure 1).

    1. 100uA are sourced from RING2 to TIP
    2. 100uA are sourced from SLEEVE to TIP

    Figure 1

    From these steps a corresponding value will be stored in the SLEEVE_Z and RING2_Z registers (Table 1).  The values in these registers will then be compared to determine what value will be stored in the GND_LOC register.  Thereupon the device will be set for Standard or OMTP.

    Table 1

    This table here gives a bit of an indication as to what impedances are expected to be seen between the rings, tip, and sleeve.  These impedances act as a voltage divider with source impedance. Larger impedance values between the tip and the sleeve will produce larger voltages seen across the tip and sleeve corresponding to your left speaker.  So for this, it seems that you have two possible options, you can either lower the resistance seen between the tip and sleeve or if you intend for this device to only use one particular audio jack, you can disable the auto detection.

    However, would it be possible for you to provide a screen shot probing from tip to sleeve?

  • In reply to Patrick Simmons:

    Kevin,

    Were you able to have resolution to this issue?

    Thank you,

    Adam

  • In reply to Adam Torma:

    Adam,

    No, but finally customer accepted this pop, thanks a lot,

    Kevin Dang