PCM1865: TDM channel configuration

Also, I would love to know if it would be possible get the PCM1807 to work in I2S format if the clocks are 256 x fs (LRCLK=fs=48000Hz, BCLK=12.288MHz). I am quite sure it is not, so I will need to change my design. At the moment I can generate the 256 x fs clocks from the PCM1865 or my DSP core, but then for the PCM1807 I do not know if I can synchronize it with the PCM1865 and get its L and R audio channels. My DSP has only one TDM input and that is why I am setting up the communications this way.

Hi Jeff,

Thanks for your answer. Now I have a problem with the configuration. I am using this audio ADC as a sensing ADC for audio purposes, but not for literal audio. I need to sense some resistors connected in the audio path so I measure the current flowing to the speakers and the applied voltage (2 speakers, so I use the 4 differential input pairs). I have checked with the oscilloscope that the signal is correctly arriving to the differential pins.

Please, correct me if I am making anything wrong in the configuration, which is:

• Set the device in master mode and configure clocks (they are fine)
• Set the output in TDM (P0, R0x0B, bits 1-0 to 0b11)
• Set TDM format to 4-ch (P0, R0x0C, bits 1-0 to 0b01)
• Set TDM words of 32 bits (P0, R0x0b, bits 3-2 to 0b00)
• Set ADC1_L to diff VIN1 (P0, R0x06, bits 5-0 to 0b010000)
• Set ADC1_R to diff VIN2 (P0, R0x07, bits 5-0 to 0b010000)
• Set ADC2_L to diff VIN4 (P0, R0x08, bits 5-0 to 0b100000)
• Set ADC2_R to diff VIN3 (P0, R0x09, bits 5-0 to 0b100000)
• Disable mic bias (P3, R0x15, bit 0 to 0)

I do not know if I have to set anything else, regarding PGA, mixers or whatever, as I do not see it clear in the datasheet. Please help, as I do see an output frame that looks like TDM, but no matter what I am inputing I cannot see a change in the read values.

Thanks in advance. BR,

Manuel

Hi Manuel,

There should be nothing to configure regarding PGA etc. Can you clarify that you see data on the DOUT pin, or are you referring to "read" as reading registers?

Best regards,

Jeff

Hi Jeff,

I mean in the DOUT pin (pin 18). The waveform that I have is the following:

• Here this is the waveform:
  
• Here you can see the cursors showing that we are using 4 channels of the TDM stream (32 BCLKs per channel)
  
• Here the cursors are showing the first channel, with the padding zeroes at the end of the word
  
• Here the cursors are showing the 24 effective BCLKs per channel, as you can infere from the difference of timing
  
• And here I have on the yellow and the purple probes the positive and negative input pins of VIN3 (pins 28 and 30), the white line is the substraction of both lines and the green line is the TDM stream, which here cannot be seen but it has no significant change
  

I hope this can be useful to you. Also, as we've been discussing in this topic, I am concerned that this may be an old unit because there is a register configuration that hints it uses the Rev.C of the datasheet. I do not know if this can make that the registers should be configured differently.

BR,

Manuel

That slope on the DOUT line is very odd. I've attached a register dump/script for you to try.

As far as the revision goes, it's highly unlikely that the part itself changed, just that the datasheet was revised to better reflect the part. 

PCM1865_Master.cfg

Best regards,

Jeff