Prevent reset by ESD event ?

Hi, Haeseong,

Excuse me, have you tried with ESD protection devices? There are some devices such the TPD1E10B09, TPD1E10B06 andTPD1E6B06 that could help to solve this issue. Additionally, you may check the System-level ESD/EMI Protection Guide. It mentions some devices that may avoid the ESD problems.

Finally, could you provide a schematic to take a look at the TLV320AIC3104 connections?

Thank you.

Best regards,
Luis Fernando Rodríguez S.

I've test with ESD diodes, but have no luck.

Please check the schematic.

Added ESD protection diode, PESD5V0S1BB, bi-directional.

Audio signal passes through FET and OP-amp, then to the CODEC input. (AINL+/-)

ESD diode doesn't make things any better, every ESD strike makes CODEC reset.

(CODEC reset pin goes low, I could see with my Oscilloscope)

Any suggesstion would be appreciate.

Hi, Haeseong,

Could you provide information about the power supplies? The power supplies must be regulated correctly with the suggested decoupling capacitors (1uF and 0.1uF). Also, they must be placed as close as possible to the codec. If it is not enough, could you try with a higher capacitance on the power pins?

Best regards,
Luis Fernando Rodríguez S.

I've tried 0.1u, 1u, 10u capacitors, but not helpful.

Here's the artwork around CODEC.

Any check point?

Hi, Haeseong,

Excuse me, could you try with the ESD components on the power pins? These devices use to be in reset condition if the power supplies are not protected correctly.

Also, could you provide the entire schematic to verify if there could be another possible cause?

Best regards,
Luis Fernando Rodríguez S.

Luis, thank you for your answer.

But I'm still in the problem. ESD protection diodes for power pins make no change.

Other same 3.3V voltage-working ICs (MCU S3F8S5A and DSP TMS320C5535) on the board are good for working,
even with ESD stike, but only TLV320AIC3104 has the reset problem.

Here's the full schematic about the CODEC.

I hope we can find any clue...

Hi, Haeseong,

Excuse me, I noticed that DVDD pin doesn't have decoupling capacitors. These capacitors are suggested for a correct behavior of the codec. Could you try placing the capacitors as close as possible to the DVDD pin?

Also, the DRVDD connection is routed only to a 10uF and 0.1uF capacitors. Each power pin must have this configuration of capacitors. So, could you add another 10uF and 0.1uF to this connection?

Thank you.

Best regards,
Luis Fernando Rodríguez S.

The datasheet and EVM board guide me to decouple between DVDD/IOVDD and DVSS.
But decoupling to AVSS is more helpful.

I added some capacitors to AVSS :

- 0.1uF between DVDD and AVSS (not DVSS)
- 1uF between IOVDD and AVSS (not DVSS)
- 0.1uF between /RESET and AVSS (not DVSS)

These pins are for digital, I know,
but anyway these capacitors make CODEC stable.

Please let me know your opinion about this,
I would do more tests...

Hi, Haeseong,

This issue should be associated with the analog side of the codec. So, placing the capacitors on AVSS instead DVSS could be helping to solve this problem.

Do you have any update on this?

Best regards,
Luis Fernando Rodríguez S.