TPA3128D2EVM: Idle output waveform

Hi Michael,

Were there any modifications done to the TPA3128D2EVM or was everything in default configuration?

Best Regards,

Robert Clifton

Hi Robert,

We didn't make any modifications to the TPA3128D2EVM. We did have the PVCC and AVCC rails shorted together and used a 24V power supply, 600kHz switching frequency. Tried both single ended and differential audio in configuration with low Idle Loss Mode selected. Loaded it with 8ohms, since the output was oscillating when unloaded.

Thanks,

Michael

Hi Michael,

That's switching noise, but it is supposed not too high.

Did you modify PVCC power decoupling caps?

Regards,

Derek

Hi Derek,

No modifications were made to the EVM.

Thanks,

Michael

Hi Derek,

Here is a screen shot of the questionable waveform on the LOUT+ node referenced to ground (while the left full-bridge channel is loaded with 8 ohms) of the TPA3128D2EVM eval board alongside an AC coupled capture of the 24V rail (PVCC and AVCC are shorted). It seems unlikely that this is caused by bad PSRR and I am doubtful that adding more bulk capacitance will solve this issue. This issue is on both the eval board and my customer's own design, and their design has >20000uF of bulk capacitance while the eval board has much less than that.

And here is a screenshot of the LOUT+ node referenced to ground while unloaded. It is oscillating at the resonant frequency of the LC output filter (~61kHz). Could this suggest a stability issue and possibly explain the previous waveform?

Thanks,

Michael

Hello Michael,

Can you tell me your input jumper configuration and can you show me your inputs(positive and negative) along with the output waveform.

Best Regards,

Luis