TAS5760M: TAS5760 Amplifier internal clock error active for 3 hours

Hi Customer,

May you help me to understand this issue in detail:

1. Do you see TAS5760 reports a clock error for a long time? Or fault pin reports that?

2. injection error, what does this mean? What actions do you do on TAS5760 (eg, I2c comannds， I2S halt or some actions?)

It would be perfect if you can describe this issue in details with scopes. 

BR，

Alix Wan

Hi Alix,

Could you share your email so that we can have conf.call in details?
looking forward to your reply.

Thanks 

For your questions,

1)Yes, from our CAN data log. 1 flag is designed for indication if internal clock error is active. The SW logic for extracting Amp register is run by 10ms. Other Amp errors(ex.DCT,OTE etc) are fine. The fault pin is not observed in CAN data and on the connector in the field.

2)Injection error meant:we intentionally made a internal clock error, ex connecting the SCLK to GND etc. then put it back to normal. The issue is gone. No action is designed when  internal clock error happened since it is non-latching error type.

Let me know if any concerns, a conf.call will be very appreciated.

Thanks

Hello,

As your analysis, clock fault is a non-latching error, so after error injection removed, AMP will go back to normal.

So still need your work to identify below:

1. during 3 hours and 15mins, the clock error is keeping reported? If yes, then you may need to check clock source problem.  If no, then please check if other faults like OC, DC, OT, are all of them are recorded?

2. clock error including MCLk and also IIC audio source error. may you check if there is possibility of IIS interface problem?

Do you have a local TI FAE contact? that will helps a lot to resolve this issue.

Dylan

Hi Dylan,

Please see my comment belows:

1. during 3 hours and 15mins, the clock error is keeping reported? If yes, then you may need to check clock source problem.  If no, then please check if other faults like OC, DC, OT, are all of them are recorded?

->Yes, keeping reported. OC,DC,OT are also recorded without errors. Since the communication between MCU and AMP works normally, that meant the I2C clock source is fine. From our clock sourcing, the original common clock source is coming from PLL. if there is clock source problems, why does it only happen in specific HW? 

2. clock error including MCLk and also IIC audio source error. may you check if there is possibility of IIS interface problem?

->After the issue happened, the sound from AMP can't be generated. Since we can't reproduce the issue, we can't check and confirm if there is a problem from I2S.

Do you have a local TI FAE contact? that will helps a lot to resolve this issue.

->There is no TI FAE available in our region. if you are from China TI team, it will be grateful to have conf.call since I am chinese native speaking. contact or info is appreciated .

If any support from my side for your support on my issue, let me know,please.

Thanks

Hello,

From the question 1's answer, clock error is keeping reported, we know it is a external clock problem.

Dylan