Hello everyone,
I am using the WM8960 audio codec chip combined with the IoT RedBoard and a Sparkfun Audio Jack to record and play back an audio signal with a delay of about one second. Using a headphone with a TRRS connector the audio is recorded with the microphone on SLEEVE and connected to LIN1 of the WM8960, along with a 2.2k Ohm resistor connected to MICB. The ADC output is send via I2S to the RedBoard to be stored in a buffer, while the data from some other part of the buffer is send via I2S to WM8960’s DAC which’s output is send to the headphone output. HPL and HPR are connected to Tip and Ring1 on the TRRS, Ring2 goes to GND and qwiik is also connected for configuration. The setup and code are almost identical to Example 8 of the WM8960 hookup guide:
The setup itself is working. I speak into the microphone and hear myself after some delay.
The problem is that I also hear an echo of myself after the same delay again, up to 3 times. (The first one can be heard although less loud, the second one is barely heard and third one if even only is recognisable with the oscillosco ). There seems to be some kind of crosstalk. If I disconnect the microphone right after tapping it I only hear the tap once. The physical microphone is far enough away from the headphones that their output should be recorded again. I suspect that some signal from the DAC is fed to the ADC or vice versa. The example doesn’t mention any noise at all.
Does anyone has any idea what the problem could be or any insight to the topic?
Kind regards