Hi
My customer is investigating TAS5805M for their next generation product.
In their experiment, it was found that the sound around 5.5kHz might behave like a notch filter with a probability of about 5%. The detail is described in below. Could you tell the customer the reason of the root cause? And please also provide them the countermeasure to prevent it and/or how to recover it when it occurs?
<Details>
1. After power on, Initial data is written through I2C just after starting LRCLK/BCLK. In this control, the phenomenon described in above is observed.
2. In their analysis, it was confirmed that the frequency of LRCLK/BCLK is not stable at start-up.
The phenomenon is prevented, if the initial data is written through I2C after waiting that LRCLK/BCLK become stable.
3. After 2. in above, if audio input source, Ext/Int, is changed several times, the phenomenon is observed.
4. The initial data can be written successfully even if the data is written during unstable LRCLK/BCLK. It has been confirmed with reading back the data.
<Questions related this>
Q1. How long time should the customer wait to start to write the initial data through I2C after LRCLK/BCLK becomes stable? Please provide the customer the required wait time, if we have.
Q2. What is the way to prevent the phenomenon?
Q3. What is the way to recover from the phenomenon?
Q4. What is the mechanism to occur this phenomenon? This is needed for the customer to understand the answer for Q2 and Q3 in above.
The following figure is the block diagram in customer's system to investigate the device.
Best Regards, Taki
