We’re having a problem with the audio levels in our system and we believe it’s a codec configuration issue.
- We’re inputting a tone to a microphone input pin, bypassing the DSP core, and then routing the audio output to the codec’s serial port output. The problem is that the digital audio level measured at the codec’s serial output port is 2.5dB lower than we expect.
- We’re inputting a digital tone to the codec’s serial input pin, bypassing the DSP core, and then routing audio to the codec’s DAC line-level output. The problem is that the line-level output from the codec’s DAC is 2.5 dB lower than we expect.
Put the two problems together and we see (2.5+2.5)=5 dB of attenuation when looping a microphone input signal out the serial port, through an external DSP, back into the codec, and out the line-level output of the DAC. The sample rate is 8 kHz.
I was wondering if there was some issue with fixed-gain settings of some of the internal components in the CODEC. Page 35 of the ADAU1772 datasheet illustrates these fixed gains. It appears that the 2.5dB should be added back into the system, but I was still wondering if our codec configuration (sample rate, codec DSP core bypass, etc) could be causing the 2.5dB not to be added back in.