• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 5 answers
  • Subscribers 31 subscribers
  • Views 298 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.

TPA3128D2: crosstalk spec

Yasushi Fujiwara
Genius 12535 points
Part Number: TPA3128D2

Hello Experts,

My customer is testing TPA3128D2EVM and have a question as follows. Would you please answer it?
[Question]
Customer measured crosstalk but it was -83dB that is too worse than datasheet spec (-100dB). Customer changed input voltage to 1Vrms then it had achieved almost -100dB. The datasheet specify the condition of crosstalk by Vo=1Vrms. Is it correct condition?

Best Regards,
Fujiwara

  • 0
    TI__Genius 9530 points

    Yasushi,

    Datasheet crosstalk was specified at the condition of 1Vrms, this is correct condition. may you confirm below;

    1. Are they using Audio Precision and measure it with the method of corsstalk?  If not what is the method?   just make sure noise floor is not accumulated into the calculation.  If not, then during calculation of  level of impacted channel / output level will be larger than expected.

    2. is there coupling between output speaker cable?   as the level is high, and long cable coupling will also degrade the crosstalk.  seperate the cables will help.

    Dylan

  • 0 in reply to Dylan Yao
    TI__Genius 12535 points

    Dylan-san,

    The 1Vrms is the output voltage level or the input voltage level, which one is you mentioned?

    Customer is using KeySight 8903B to measure the crosstalk. I think it can be measure the crosstalk.
    [measurement method]
    1. Adjust the input signal level(Rch) to the output voltage is 1Vrms.
    2. Adjust the input signal level(Lch) to the output voltage is 1Vrms.
    3. Rch output level is set 0dB, then Rch input is disabled(off). Lch input is enabling during measurement.
    4. Measure Rch output level. Lch output level - Rch output level = crosstalk level.

    I would like to know noise filter setting with the audio precision. What noise filter did you use while crosstalk measurement?

    Best Regards,
    Fujiwara


  • 0 in reply to Yasushi Fujiwara
    TI__Mastermind 19410 points

    Hi Fujiwara,

    1Vrms is output level.

    We suggest customer use AP instrument to measure crosstalk performance, and make sure the noise level can be correlated to datasheet spec (65uV, table 6.6 in datasheet).

    Regards,

    Derek

  • 0 in reply to Yanming Gong
    TI__Genius 12535 points

    Hi Derek-san,

    The noise level is 67uVrms. My understanding of the crosstalk measurement is;
    1. Measure output noise level for each channels.
    2. Adjust the input signal level(Rch) to the output voltage is 1Vrms.
    3. Adjust the input signal level(Lch) to the output voltage is 1Vrms.
    4. Rch output level is set 0dB, then Rch input is disabled(off). Lch input is enabling during measurement.
    5. Measure Rch output level. Lch output level - Rch output level - output noise level(65uVrms) = crosstalk level.

    Is my understanding correct?

    Best Regards,
    Fujiwara

  • 0 in reply to Yasushi Fujiwara
    TI__Mastermind 19410 points

    Hi Fujiwara san,

    For item #5, since Rch has no audio input, so Rch output is almost output noise. So you can check 20log(Rch/Lch).

    Regards,

    Derek

  • 0 in reply to Yanming Gong
    TI__Genius 12535 points

    Hi Derek-san,

    The -100dB is 10uVrms if output voltage of Lch is set 1Vrms. The output noise level is 65uVrms(typ), so 20LOG(Rch/Lch) is -83.7dB. Does customer measurement result reasonable?

    Best Regards,
    Fujiwara

  • 0 in reply to Yasushi Fujiwara
    TI__Mastermind 19410 points

    Hi Fujiwara san,

    I am sorry that datasheet data is based on 1W output power, not 1Vrms output. 

    And for crosstalk calculation, the noise should be value at 1KHz, not a-weighting noise.

    So we suggest they can use AP instrument to calculate performance correctly.

    Thanks!

    Regards

    Derek

  • 0 in reply to Yanming Gong
    TI__Genius 12535 points

    Hi Derek-san,

    Would you please tell the schedule when you modify the datasheet?

    BTW, AP will calculate the noise@1KHz automatically when it is measured the crosstalk, correct? If possible, would you please tell noise density?

    Best Regards,
    Fujiwara

  • 0 in reply to Yasushi Fujiwara
    TI__Mastermind 19410 points

    Hi Fujiwara san,

    I think it is not necessary to modify datasheet now. Please tell customer the datasheet data is based on 1W output.

    AP instrument has crosstalk test item, and select the crosstalk test item, the AP instrument will test crosstalk automatically.

     And datasheet data is based on EVM board.

    Regards,

    Derek