While making a sigmastudio file on a ADAU1401 system with external ADC and DAC, with the DSP as Master I found a very strange phenomenon. Included are 2 screenshot showing my design with the sigmastudio set at 48kHz and 96kHz. When switching from 48 to 96 or vice versa, I also change the Register control program length/Frame sync frequency and BLCK frequency to the appropriate settings as described in a post on the forum by Brett I think.
On both screenshot I have a 10kHz sinewave feed into the external ADC. The problem maintains if I change frequency or level of input. On the 48kHz file you can see that the level bars are showing normal levels through the whole design. On the Dig0 channel the level bars show almost no signal because there is a lowpass filter on it at 1300hz. On the Dig1 channel you can see the level bars showing signal and also after the addition of 2 signal paths the level does not comes into clipping territory.
On the 96kHz file however, you see that after the add block of both channels the signal levels are gone crazy, constantly clipping into 0dB level. Also the DSP won't output any signal on the Sdata out into the external DAC anymore. It does not matter at all what kind of add function block I use (mixer, signal merger, etc) after the mixing point of 2 signal paths the DSP is going banana's....
I really want to run on 96kHz, which should be possible I think, but I have no clue what is causing this issue. It seems like a bug in the SIGMAstudio to me. Maybe in combination with the ADAU1401/1701 chip, I don't know as I only have a test boards with this chip.
Anyone got a clue or a solution to this issue? I rather have a hurry to solve this as I need these systems in production soon...