I'm having a very strange behaviour of ADC1 on an Adau1701 standalone board that I designed. The 1701 runs in MASTER mode, so it generates the clocks (BCLK and LRCLK). The MP3 pin of the chip is used as an Aux ADC, and is connected (according to the application notes) to the center pin of a 10K pot (between 3.3 and GND) with a 475 Ohm resistor.
What happens is that when the Mp3 pin is configured as ADC1 the system becomes unstable, and depending on the position of the POT the LRCLK actually switches between 96K to 48K (and noises are generated at the outputs).
Initially I thought about a hardware mistake, or some soldering fault. After many checks I realized that everything is OK. I replicated the same behaviour on a different board. I also use ADC2 (on pin MP2) wired in exactly the same way to another pot, and everything is OK.
I tought about a possible short between the MP3 pin and its adjacent pins ... but no, it's all OK.
In the actual situation I can't use the ADC1. The rest of the board works OK.
But how can it be? Does anybody happened to see something similar on their projects?
How is it possible that a pot connected to ADC1 actually controls the sampling rate in a standalone application?
Can the chip be faulty?