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.

TLV320AIC3106 - noise and strong ultrasonic component at LINE and HP outputs

Other Parts Discussed in Thread: TLV320AIC3106

I have encountered an issue with TLV320AIC3106 in a newly developed custom board with an ARM processor, Linux 3.0.35 and driver tlv320aic3x.

Both LINE and HP outputs exhibit quite a large amount of audible noise. The noise sounds similar to a cassete player noise but is a deal larger. As to my ears, sound itself is reproduced accurately. Noise appears AFTER the DAC is enabled but is not affected by muting the output mixers.

Perhaps the noise origins from a strong ultrasonic comonent at the outputs, which appears at the same time. The component is ~200mVp-p ~530kHz and is OF THE SAME PHASE at both terminals of the differential output port. (attached are the waveforms at the LINE output after a 1st order RC-filter at the input of a D-class amplifier). The component is not dependant on the sampling frequency (44.1/48kHz), master clock frequency (24/12MHz) and master/slave clocking mode, but shows some dependency on the programmed common-mode voltage: it gets a bit lower in amplitude and a bit higher in frequency when VCM changes from 1.35V to 1.8V. Meanwhile the common-mode voltage at the inputs is free of such a disaster. Two boards have been checked to behave similarly.

Power lines show the traces of the same frequency ~10mVp-p. Adding/removing capacitors/ferrite beads to them has no visible effect on the ultrasonic component.

Any help is much appreciated.

Sergey

TLV320AIC3106_Noise.tif
  • Hi, Sergey,

    I think you are seeing issues with the out of band noise from the sigma-delta converters using in the TLV320AIC3106.

    Please try inserting an RC LPF with a cut-off frequency of around 30 kHz between the codec and your power amp, and I think you will find the noise will be greatly improved.

    -d2

  • Hi, Don,

    In fact the issue happened to be caused by some mystical interaction of the codec' LINE-R output with the input of an onboard GSM modem from Cinterion. Anyway after the LPF is inserted into THIS path, ALL the codec' outputs are doing well.

    I think the issue can be closed.

    Thanks,

    Sergey.