• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 4918 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.

Dual microphone, noise cancellation method

Seng Huat Toh
Prodigy 60 points
Other Parts Discussed in Thread: LP2966

Hi,

I had made an attempt trying simulate a dual microphone, noise cancellation method, with the Supplies to this circuit is 3.3V and GND.

Below is the snapshot of my circuit, I had also attached my schematic file 7762.PRENOISwSUMAMP_BPF_COM.TSCof my circuit.

What I'm not sure from my circuit is the voltage offset for the op-amps. From past experience, the input offset voltage get amplified at the output and the desired signals get saturated. However, as i simulate this circuit, it's not the case, the output stays at the offset voltage level. So 'im confused here.

Please kindly advise what I should take note of, whenever the op-amp circuit supplies are not dual but V+ connects to 3.3V and V- is to GND.

Awaits reply

Regards

Seng Huat

  • 0
    TI__Guru**** 189399 points

    Hi, Seng Huat,

    I didn't simulate your circuit, but by reviewing your schematic, you have stages ac-coupled with capacitors which will eliminate dc offset errors, won't they?

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 60 points

    Hi Don,

    Yes, at the stage with the last two op-amps, it is for the filtering purpose (2nd order band pass) this had block off the offset voltage, as such I had a 1.5M ohm resistor pull up (1.6V) right before the input.

    At the very last stage, I had followed the recommended circuit from the datasheet, to place a capacitor (dc offset stage) before the analog microphone input of an audio codec.

    Working on the op-amp circuit, I'm limited to 3.3V and GND (0V), so for an op-amp circuit, I know I need to have a offset voltage for the op-amps input so that the input signals won't get cut-off or saturated. Since I only have one supply (3.3V), my only way to 'make up' the offset voltage is through a voltage divider. What I don't understand is that what I can do so that I can obtain the same voltage offset level at every op-amp stage.

    Hope I can clear my doubts in this conversation

    Regards

    Seng Huat

  • 0 in reply to Seng Huat Toh
    TI__Guru**** 189399 points

    Hi, Seng Huat,

    There are a couple ways you can generate this voltage. We usually refer to this as Bypass or Mid-rail voltage.

    In your case, with 3.3V supplies, you would want to target 1.65V as your mid-rail voltage.

    You can generate this with a simple resistor divider, but any variations on your supply will couple thru to the audio signal.

    Or, you can use a precision reference to give you exactly 1.65V.

    -d2

  • 0 in reply to Don Dapkus
    Prodigy 60 points

    Hi Don,

    Thanks for the respond.

    I had my schematic being reviewed by my EE team in my company actually, so the feedback summary is my circuit is too complicated... I was then recommended to try to use a Differential amplifier configuration instead.

     

    Infact from this simulation, I had actually get my desired result such that the noise source get cancelled out at Vout. This is a simulation under the scenario that The amplitude of the noise source (which is a white noise wav file) is 2x larger than the noise source from the voice source.

    Like you said, I need another voltage source for the offset voltage i had chose LP2966 (1.8 and 3.3) already, thanks for sharing the TI-Burr-Brown part with me.

    I'm going to build the prototype circuit, so I may be I'll share my experiment results later here, since I started this problem..

    Regards

    Seng Huat