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.

PCM1820: What is the bias voltage of the audio input?

Part Number: PCM1820
Other Parts Discussed in Thread: PCM1862, PCM1860, OPA1692
Dear support team,

I confirmed that and thought that the bias voltage was around 1.5 ~ 1.45V.
Seems to be sufficient for full-scale input 2Vrms.
The supply voltage of my system is 3.3VThe amplifier in the previous block is also 3.3V.That means that the amplifier output is up to 3.3V.
When a 3.3V signal enters the PCM1820, it first exceeds the specification of negative voltage limitation.
I suspect that the bias is not half the power supply. Since there is a 3.3V input under the worst conditions in the system, I would like to avoid breaking the PCM1820 low voltage limitation by halving the bias of the power supply.
Is there a way to do that?

Best regards,
Yukio Oyama
  • I made a mistake in the last part.
    Correctly,
    I would like to avoid breaking the PCM1820 low voltage limitation by the bias to half of the power supply. Is it possible to shift the bias

  • Hello Yukio,

    I am not sure what negative voltage limitation you are talking about but everything on the chip has voltage controlled by regulators that cannot be changed.

    What exactly is the issue you are running into?

    Best Regards,

    Low Power Audio Apps Team

  • Dear Low Power Audio Apps Team,

    It is a limit due to the absolute maximum rating. The lower limit of the analog input voltage is -0.3V in the datasheet.

    In our system, the input voltage from the front block to the analog input is 3.3Vpp at worst. The PCM1820's analog input bias is probably lower than AVDD / 2 (= 1.65V), so the 3.3V input voltage will cause the lower bound to break first.

    Does your answer mean that you can't adjust the bias voltage of the analog input voltage?

    Is it impossible to change even if an external circuit is added?

    I want to use up the full scale.

    However, the full-scale voltage is only described as a typical value. If the front block output is restricted to 2Vrms, it will not reach full scale depending on device variations.

    Being below full scale is not good than exceeding it, as it affects our product specifications.

    Best Regards,

    Yukio Oyama

  • Hello Yukio,

    Could you clarify whether or not you are trying to use as single-ended or differential input?

    2Vrms full-scale will only be reached via differential input on this ADC, 3.3Vpp is max you can put on each pin. If you put 3.3Vpp on minus and plus you will get full-scale.

    We do not recommended self-bias as the specs of datasheet are produced via AC-coupling.

    We have other ADC products that include programmable gain before ADC for situations where full-scale is exceeded but not this device.

    Best Regards,

    Carson

    Low Power Audio Applications

  • Hello, Carson-sann,

    I use it with a differential input in the ADC. I understand that 2Vrms is the value at the differential input. In our design, I want to use up the full scale of the ADC, so I apply 3Vpp to each of the INxP/M analog input ports. But, when the user inputs more than the product specifications, the waveform is clipped to the power supply at the front block and 3.3Vpp is applied to the ADC. At this time, the minimum voltage value is below the absolute maximum rating.

    Which ADC has programmable gain? Is the bias voltage the center of the voltage?

    Best Regards, 

    Yukio Oyama

  • Hello Yukio,

    A few questions to understand your question better:

    1. Are you using AC-Coupling?

    2. Do you have DRE enabled?

    3. Can you send me your schematic?

    The PCM1860 or PCM1862 both have programable gain.

    Best Regards,

    Carson

  • Hi, Carson-sann,

    The answer to each question.

    1. Yes, we are coupling with a 22uF capacitor.

    2. No, we are using another method in our signal processing. Using the device's DRE is confusing to our signal processing.

    3. No I can't. There are many barriers to exposing schematics to the outside.

    However, it is possible to give an overview.

    The amplifier circuit of the front block is a differential amplifier circuit that is often used in the first stage of instrumentation amplifiers. The OPAMP uses your OPA1692 and its supply voltage is 3.3V.

    Thank you for introducing your device.

    However, the dynamic range of PCM1860/1862 is not enough for us. The good thing about PCM1860/1862 is that the full-scale voltage input is fixed at the maximum value.

    Can the PCM1820 also specify the maximum full-scale voltage input? When specified by typical values, we are concerned that the analog input voltage will be less than digital full scale. Therefore, we set an analog input voltage that is slightly larger than the typical value. However, it causes the problem of this case that the ADC cannot be electrically protected against excessive input to the analog front end. By adding many circuits, It is possible to  protect it,  but it is not cost-effective.

    Best Regards,

    Yukio Oyama

  • Hello Yukio,

    I will need to confer with systems team and get back to you on this.

    Best Regards,

    Carson

  • Hello Yukio,

    You cannot specify full-scale like you would want with this device, best way to approach is to look look at 2Vrms as Vmax.  if you are worried that does not bring you to full-scale then look at Vref and verify it adheres to this paragraph from datasheet.

    DRE was designed to get max dynamic range of 123dB but without it you have that reduced 113dB.

    If you are worried about electrically damaging the device then either keep at or below 2Vrms or use DRE, otherwise we cannot guarantee.

    Best Regards,

    Carson

  • Hello Carson-sann,

    I understand that you cannot specify the maximum value of the standard.

    I will decide it statistically or consider a different device.
    Thank you for your advice.

    Best Regards,

    Yukio Oyama