• Resolved

TLV320AIC3254: CC3200AUDBOOST POP Noise

Genius 4315 points

Replies: 24

Views: 425

Part Number: TLV320AIC3254

Hi

We use "LAUNCHXL-CC26X2R1"+ "CC3200AUDBOOST"  to evaluate our project.

We found / observed what the headphone-out of CC3200AUDBOOST has the POP noise when it is powered up.

Do we need to follow up "2.2.3.1 Stereo Single Ended Configuration" to add the Schottky diode to CC3200AUDBOOST?

If yes, do we need to change AIC3254 configuration? please refer to the following our AIC3254 setting:

// Configure Power Supplies
AudioCodec_pageSelect(TI3254_PAGE_1);        //Select Page 1
AudioCodec_regWrite(TI3254_PWR_CTRL_REG, 0x08);  // Disabled weak connection of AVDD with DVDD
AudioCodec_regWrite(TI3254_LDO_CTRL_REG, 0x01);  // Over Current detected for AVDD LDO
AudioCodec_regWrite(TI3254_ANALOG_IP_QCHRG_CTRL_REG, 0x32); // Analog inputs power up time is 6.4 ms
AudioCodec_regWrite(TI3254_REF_PWR_UP_CTRL_REG, 0x01);   // Reference will power up in 40ms when analog blocks are powered up

//De-pop: 5 time constants, 6k resistance
AudioCodec_regWrite(TI3254_HP_DRV_START_UP_CTRL_REG, 0x25);  // Headphone ramps power up time is determined with 6k resistance,
                   

Thanks.

BR

Trevor

  • Hi Trevor,

    If you have concerns on pop during system power-up, then please follow the recommendations in  http://www.ti.com/lit/slaa492a.

    Pop during HP power-up can be minimized by selecting the right Rpop value ("2.2.3.1 Stereo Single Ended Configuration") and by using a proper power up sequence in the driver (refer to attached file).

    The snippet that you have shown is only partial. After setting the Rpop value, power up the DAC and HP amps in the recommended sequence.

    Best Regards.output power up sequence

  • In reply to Diljith:

    Hi Diljith,

    If we need to add the Schottky diode(http://www.ti.com/lit/slaa492a.)

    Do we need to modify the following register?

    AudioCodec_regWrite(TI3254_PWR_CTRL_REG, 0x08);  // Disabled weak connection of AVDD with DVDD
    AudioCodec_regWrite(TI3254_LDO_CTRL_REG, 0x01);  // Over Current detected for AVDD LDO

    BR

    Trevor

  • In reply to Trevor Hsu:

    Yes, TI3254_PWR_CTRL_REG can be set to 0x08 and TI3254_LDO_CTRL_REG can be set to 0x01. Always select AVDD LDO output >= DVDD LDO output.

    Best Regards.

  • In reply to Diljith:

    Hi Diljith,

    Regarding LDO Control Register(P1_R2), current we set to 0x01(00: DVDD LDO output is nominally 1.72V , AVDD LDO output is nominally 1.72V), but we measured the the voltage level, the AVDD = 1.683V and the DVDD = 1.69V.

    we also measure the sequence for AVDD & DVDD, please see the following figure(the blue line is AVDD and the yellow line is DVDD):

    # Without Schottky diode

    # With  Schottky diode

    Do you have any suggestions to us?

    Thanks.

    BR

    Trervor

  • In reply to Trevor Hsu:

    Hi Diljith,

    Update some information for you.

    Please see the following sequence for internal LDO. we found that the POP noise occurs in the yellow circle.

    Do you have any suggestions to us?

    
    

    BR

    Trevor

  • In reply to Trevor Hsu:

    Hi Diljith,

    Regarding your headphone out sequence, the pop noise occurs in the following line :

    # Power up HPL/HPR and LOL/LOR drivers

    w 30 09 3C

    Do you have any suggestions?

    BR

    Trevor

  • In reply to Trevor Hsu:

    Hi Diljith,

    We also checked the POP noise on the AIC3254 EVM, there are four POP noises on EVM.

    The first one is to plug USB cable into EVM(Power on codec). The second one and third one are happen when using AIC3254 CS to load "High Performance Stereo Playback". The last one is unplug the USB cable from EVM(Power off codec).

    We also conducted some experiments:

    1. Regarding 2nd POP, we found the 2st POP is related P1R9.

    2. Regarding 3rd POP, it is related to P1R20:

        If set to 2k resistance, 2nd POP & 3rd POP are very close

        If set to 25k resistance, 2nd POP & 3rd POP are far away.

        BTW, regarding 3rd POP, it only happens when a silent source is played.

    Do you have any suggestions for us?

    BR

    Trevor

  • In reply to Trevor Hsu:

    Hi Diljith,

    Do you have any update?

    BR

    Trevor

  • In reply to Trevor Hsu:

    Hi Trevor,

    We can only minimize the pop through the de-pop register settings and sequencing the power-up of various elements in the signal chain; we cannot eliminate it. If you try to remove the de-pop setting (P1R20), you will find that the power-on pop (P1R9) will be much louder. So the overall strategy of the register writes that I shared is to minimize the pop. Would this level of pop be a matter of concern for your application?

    I reviewed the results with and without the Schottky diode and I can see that the diode is supplying the voltage to AVDD after DVDD is powered up. So there should be a reduction in the pop. Were you able to measure pop with and without the diode?

    Best Regards.

  • In reply to Diljith:

    Hi Diljith,

    As previous state, there are four POP noises on the EVM, could you observe this result?

    Regarding remove the de-pop setting (P1R20), the 2nd & 3rd pop will be mixed together. so the pop will be much louder.

    Currently, we are focusing on 1st & 4th pop noise.

    Currently, we are focusing on the first and fourth popular noises. Regarding these two pop noises, I think it is related to power sequence.

    As I know the EVM uses external DVDD & AVDD power supply and the CC3200AUDBOOST uses the internal DLDO and ALDO power supply. 

    Regarding the Schottky diode, it should apply to internal DLDO and ALDO power supply. I don't understand why the EVM need to add the Schottky diode.

    Anyway, could you provide your AVDD and DVDD sequence to us(waveform)? Could you also provide the your Schottky diode(part number / DS)?

    About 4th pop noise, could you provide the register setting for power off?

    
    

    Thanks.

    BR

    Trevor