PCM6020-Q1: from ADC(PCM6020-Q1) can read voltage value?

Hi Lawrence,

The raw voltage data read of MICBIAS, VBAT_IN, and input channels can be read in page 1 diagnostic registers (0x5A-0x65).

The equation to convert the 12-bit value to decimal:

Explained here:  PCM6xx0-Q1 Fault Diagnostic Features

Regards,

Hi Daveon:

Does this voltage data only support DC-coupled input?

Because when I set CH1 as AC-coupled input in the P0_R60 register, the raw voltage data for IN1P and IN1M is 0.

However, when I set CH1 as DC-coupled input in the P0_R60 register, the raw voltage data for IN1P and IN1M is not 0.

Is there a method to read the raw voltage data if I set CH1 as AC-coupled input? 

The ADC CH1 input voltage is amplifier output. Please refer to the image above.

Thanks,

Leo

Hi Leo,

The voltage data along with input-fault diagnostics is only supported when the device is DC coupled.

Hi Daveon:

In the original question, it was suggested to employ an AC-coupled design.

If we were to change to DC-coupled, would there be any concerns?"

Hi Leo,

You can use diagnostics with an AC coupled configuration, however you would sacrifice the use of some input channels. Example shown below:

\

Whether AC or DC coupled, the ADC must be biased to common mode. Please read section 8.3.3 of the d/s for a detailed explanation on how to configure the inputs for its respective configuration. Resistor choice guidance is given here as well.

Regards,

Hi, Daveon:

Do you mean I have to set up MICBIAS regardless of whether it is AC or DC coupled?
Does the MICBIAS affect the amplifier output voltage if MICBIAS and the amplifier output are connected, as shown in the diagram?

Thanks,

Leo

Hi Leo,

If you are using an ECM microphone, in a DC coupled configuration, you can use the MICBIAS to bring the input signal of the MIC to the common mode of the ADC input. When AC-coupled, DC voltage is blocked, so depending on your MICBIAS level you still need to add resistors to bias the input signal level to common mode. A differential line input, whose common mode difference is 0V, would not need biasing unless negative voltages are sent.

However, in this case the output is coming from a (speaker?) amplifier. Similar to linked post, I recommend to AC couple the output of the AVAS amplifier to PCM6020. Otherwise, yes, the MICBIAS would affect the output signal level by solely adding a DC bias.

Diagnostics such as voltage read are only available when DC coupled, so customer would have to use a separate channel for diagnostics 

Regards,

Hi, Daveon:

I connected the amplifier output to the ADC CH1 and CH2 inputs. CH1 is set to AC-coupled, and CH2 is set to DC-coupled. However, when I enable the diagnostic function for both channels, it detects faults and triggers an interrupt. Upon checking the register, I found the cause: "input short to each other detected" on CH2. I also tried to mask the interrupt for CH2 in the INT_MASK2 register, but it didn't work.

Do you have any suggestions to avoid the diagnostic function triggering an interrupt when it detects an "input short to each other" fault?

Thanks.

Leo.

Hi Leo,

This fault triggers when the absolute value of the difference between the input pins falls under the programmed threshold. It is programmable from 0 V–450 mV in steps of 30 mV. I suggest changing this threshold to a higher value.

Regards,

Hi Daveon:

I connected the amplifier(TAS6422-Q1) output to ADC input channels 1 and 2, with channel 2 used for DC diagnostics, as shown in the figure below. I obtained the voltage values from the ADC IN2P and IN2M data registers. By averaging the positive and negative values (i.e., (IN2P voltage data + IN2M voltage data) / 2), Can this "averaged voltage" be used to represent the amplifier's output voltage during sound playback and to calculate the amplifier's output power?

Thanks.

Leo

Hi, Daveon

I would like to clarify some information. Would you recommend using RMS voltage rather than average voltage for IN2P and IN2M?

Additionally, should the RMS voltage be calculated separately for IN2P and IN2M, or should they be combined to determine a single RMS voltage?

Thanks.

Leo

Hi Leo,

You can multiply Vavg * 1.11 to find Vrms or through the equations below.

 Image Source: https://www.electronics-tutorials.ws/accircuits/rms-voltage.html 

If the signal is differential, you have two of the same signal 180 degrees out of phase with each other. To simplify, you can compute your calculations with just INxP.

Regards,