• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 2520 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.

TAS5558: Output noise when muted

Freddie Temperton29
Prodigy 100 points
Part Number: TAS5558
Other Parts Discussed in Thread: TAS5624A, DIR9001, PCM1808, SN74LVC2G157

I have the TAS5558 connected to a TAS5624A power stage and am experiencing noise on the output of the TAS5558 when there is no input, even when the hard mute IO is pulled low too.

I have eliminated the power stage (the TAS5624A) as the suspect, as I can change the clock inputs to the TAS5558 from 44.1 to 48Khz sampling rate, and you can hear an audible change in pitch from the speaker on the output.

Does anyone know how to stop the TAS5558 producing this noise? It is clearly violating all it's SNR ratings in the datasheet! I feel it could be something to do with this oscillator trim register... I'm setting register 0x12 to 0x00 to get the chip to even work

Thanks in advance

Freddie

  • 0
    TI__Guru 61280 points
    Freddie,

    Setting 0x12 to 0x00 is the standard bring up procedure, you're doing nothing special there.

    As for your noise issue, is this on the EVM? What power supply are you using? Have you observed the noise on a scope? If so please send me an image.

    Regards,

    -Adam
  • 0 in reply to Adam Sidelsky
    Prodigy 100 points

    Adam,

    Thank you for getting back to me. I have just conducted some experiments and taken some FFTs to share.

    This is not on the EVM but my own implementation. The schematics are heavily influenced by the EVM though so I would find it unlikely my implementation is to blame, but I'm keeping my opinions open for now!

    It should be noted that both the tests below are performed with the TAS5558 muted! (ok, only soft mute via the mute input however)

    The exact scenario for the worst noise (massive peak at just under 16kHz) is 48kHz LRCLK, 3.1MHz BCLK and 12.3MHz MCLK (256fs oversampling and 32bit)

    If however I change the input to 44.1kHz LRCLK, 2.82MHz BCLK, 11.2MHz MCLK the 16kHz spike goes down, however there are quite a few others with there being audible broadband noise.

  • 0 in reply to Freddie Temperton29
    Prodigy 100 points

    And as for the power supply, I'm using a 12V 5A rated bench power supply with >2000uF on-board capacitance which the 3.3V for the TAS5558 is derived from, and another couple hundred uF on the TAS5558 3.3V rail. The noise is also observed equally on both channels so I don't think it is a board routing pickup type issue with the power stage

  • 0 in reply to Freddie Temperton29
    TI__Guru 61280 points
    Freddie,

    I see some potential misalignment with your signals:

    "48kHz LRCLK, 3.1MHz BCLK and 12.3MHz MCLK"

    Your BCLK is 64.58333 multiple of your LRCLK, can you align this as 64 exactly?

    Your MCLK is also 256.25 multiple of LRCLK, can you align this as 256 exactly?

    Lining up these ratios to whole integers may help. Sometimes the part can handle variation like this but in certain cases the differences can cause noise.

    Regards,

    -Adam
  • 0 in reply to Adam Sidelsky
    Prodigy 100 points

    Hi Adam

    Thanks for your reply. The numbers I quotes were just rounded approximations - I have confirmed the clocks are exact 64 and 256 multiples. The noise is still present

    Thanks

    Freddie

  • 0 in reply to Freddie Temperton29
    TI__Guru 61280 points

    Freddie,

    I am not sure what is causing this. Please send me your schematic and/or layout and I'll take a look.

    Regards,

    -Adam

  • 0 in reply to Adam Sidelsky
    Prodigy 100 points

    HI Adam

    Here's the schematic for the power stage. The data is coming from either a DIR9001 or PCM1808 and muxed through SN74LVC2G157

    Freddie

  • 0 in reply to Freddie Temperton29
    TI__Guru 61280 points
    Freddie,

    Sorry for the delay, were you able to resolve the issue?

    I don't see issues in the schematic, can you share the layout?

    Regards,

    -Adam
  • 0 in reply to Adam Sidelsky
    Prodigy 100 points

    Hi Adam

    No luck with the noise issue. I have discovered that when the amp is first turned on and warming up, the dominant frequency tends to drift around a lot. Very strange.

    Attached are the gerbers for the board

    Freddie

    hifi_amp.zip

  • 0 in reply to Freddie Temperton29
    TI__Guru 61280 points

    Freddie,

    I thought of something we can try. Use the TAS5558's AM avoidance to check if some clocking or interference issue is causing the problem. 

    See Datasheet section 7.6.2.43 and register 0xDE for details. If something on the board is beating with some internals of the modulator, it could cause noise. 

    Regards,

    -Adam