• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 31 subscribers
  • Views 1494 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TAS5760 - high output noise and 11kHz noise

Vio Liu
Expert 3045 points

Hi team,

I'm Vio Liu from Guangzhou, China. My customer is using TAS5760. Currently they are facing two questions. 

1. Large output noise. For TAS5760, the noise on datasheet is 100uVrms. While for other amplifiers, it is usually 40~60uVrms. TAS5754/TAS5756/TAS5766 has smaller noise as well. My customer is using TAS5760 and they have audible noise, which is about 120~140uVrms(A-weighting). Do we have any way to reduce the noise?

2. If we short the DIN pin of I2S to ground, we can see a 11kHz noise on the output. The same thing also happened on TAS5760EVM shang TAS5756 EVM. If the input signal is not always zero, there's no peak on 11kHz. Could you please let me know why? Thanks.

Best regards,

Vio Liu

  • 0
    TI__Mastermind 20085 points
    Hi Vio

    Thank you for your question. My colleague will answer you soon with further information.

    Best Regards
    Jose Luis Figueroa
    Audio Applications Engineer | A-Team
  • 0
    TI__Guru 57800 points

    Hi, Vio,

    The noise level on the TAS5760 is directly related to the gain applied in the amplifier. Changing the gain structure might help reducing the noise in the amplifier, can you please try following the steps in section 9.4.1.7.2 (hardware mode) or 9.4.2.4.2 (SW mode) from the datasheet to try reducing the noise in the system?

    Can you please provide instructions on how to replicate the 11KHz noise issue in the EVM?, like configuration of the amplifier, input used, etc. 

    Thanks and Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Diego Melendez
    TI__Expert 3045 points
    Hi Diego,

    1. Do you mean the 120uVrms noise is reasonable when gain is set to 25dB?

    2. If we short the Din pin of TAS5760 to ground on our EVM, then connect the EVM to computer with USB cable, open PPC2. After initializing, we can get the FFT graph with 11kHz peak. If remove the short DIN, then 11kHz peak will disappear.

    Thanks.
    Best regards,
    Vio Liu
  • 0 in reply to Vio Liu
    TI__Guru 57800 points

    Hi, Vio, 

    Please refer to my comments below:

    1 - Yes, the noise spec is about 120µVrms with 25dB gain set in the amplifier if power supply is 24V, however, if the gain of the amplifier is not properly configured, the noise can be this high with lower power supply level. 

    2 - I followed the procedure you mention and found a similar peak. I also found that this peak is only present when DIN is shorted to GND, after removing the short, the peak disappears. I tested the EVM without shorting DIN with GND before connecting with PC and found there is no peak in the FFT plot. Can you replicate this?. Also, I would like to point out that the 11KHz peak level is under the amplifier's spec. 

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Diego Melendez
    TI__Expert 3045 points
    Hi Diego,

    Thanks for your great help.
    1. Since my customer has chosen TAS5760 as the audio amplifier, do you have any method to reduce the audible noise? If there's some documents or application notes, it would be ok as well.
    2. Yes, I can replicate. Also it is totally the same with my test result. We are curious about why could this happen, even though it is under the ampilfier's spec. Also customer is interested in whether this could cause any potential risk.


    Thanks.
    Best regards,
    Vio Liu
  • 0 in reply to Vio Liu
    TI__Guru 57800 points

    Hi, Vio, 

    As I mentioned before, configuring the gain of the amplifier as directed in sections 9.4.1.7.1 and 9.4.2.4.1 of the datasheet will help reducing noise level. Basically, you need to configure the amplifier with the lowest gain that allows the user to achieve the desired output power into a determined load. The peak might be caused by noise coupling into the system's ground when shorting DIN to GND. There should be no potential risk as the peak is under the amplifier's noise spec.

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Diego Melendez
    TI__Expert 3045 points
    Hi Diego,

    SDIN pin is weak pull down inside TAS5760. If there's low frequency noise such as 11kHz coupling into ground, the noise should have strong drive capacity. Why there's strong drive capacity for the SDIN pin?

    Best regards,
    Vio Liu