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.

TLV320ADC3101: IIR filter and biquad filter performance

Part Number: TLV320ADC3101

Hi, 

According to:  https://e2e.ti.com/support/audio-group/audio/f/audio-forum/1010866/tlv320adc3101-need-suggestion-to-reduce-dc-offset-of-tlv320adc3101-to-zero/3745330?tisearch=e2e-sitesearch&keymatch=TLV320ADC3101%252520filter#3745330

Why does IIR filter has better performance than biquad filter for eliminating the DC component?

The equations on the datasheet are below.

Is biquad equals to 2nd order IIR? And more orders gives better eliminating performance? 

 

Best regards,

Angela

  • Hi Angela,

    Either filter can work as low-pass filter to take out DC just as well as the other

    In other post they were basing what filter would be best for what based on that customer's coefficients they were using.

    2nd order does usually give better cutoff performance but in either case if z =0 then answer comes out to be same for each.

    Best Regards,

    Carson

  • Hi Carson,

    We have idle tone problem for differential input. Adding dither control eliminate idle tone, but increase the DC bias.

    We try two filters to reduce DC bias, IIR filter and first order biquad filter.

    • IIR filter register: N0: 7F D5, N1: 80 2B, D1: 7F AA
    • First order biquad filter register: N0: 7F D5, N1: C0 15, N2: 00 00, D0: 3F D5, D1: 00 00

    These two filters have similar transfer function H(z). However,  IIR filter has better performance than first order biquad filter to reduce DC offset.

    Thank you.
     Angela

  • Hello Angela,

    Thanks for your patience, but could you be more specific when you say better performance?

    Best Regards,

    Carson

  • Hi Carson,

    We are measuring microphone noise floor. DC component is something we don't want.

    IIR filter have better performance eliminating DC component.

    Thank you


    Angela

  • Hello Angela,

    I see now. Could you do me a favor and make it so formulas identical by making N1(IIR) = 2 x N1(BIquad), D1(IIR) = 2 x D1(Biquad)

    and run that same frequency sweep and tell me if there is any change.

    Thanks,

    Carson