testing FxNLMS block and Switch On/Off block in a real time ANC tube

Hello Alessandro2012,

Can you describe these "low frequency divergences" in a little more detail?

Do you get oscillations? Poor cancellation performance?

Since it is in a tube I would expect the resonate frequencies should not shift much if any? Then is it the source of the noise that is changing? 

Are you seeing the filter coefficients changing when the adaptation is shut off? It is not supposed to do that. 

We may need to see your project but if you can describe the problem a bit more it will help us to understand what is happening. 

Dave T

Added Edit:

Since the time zones are very different between us, can you also send a lot more information? It would be great to see a picture of your setup. Where the mic is and noise cancelling speaker in relation to the tube and noise source. The position of the mic in relation to the speaker and noise source is important. What is the sampling rate on your project? This will determine the latency. Do you fully understand the noise path delays?

This may be more info than you want to provide on the forum and I understand that. So send me a PM if you want but you must also send a friend request or I will not be able to respond. 

Dave T

Hi Dave, 

could you test the delay of the ADAU1452 considering the analog input and analog output? (simply considering the connection between the sdata input block and the output block as a project in sigmastudio)

thank you