TPA3255: Clipping Pulse Injector and output Waveform ringing even with no load

Hi,

You can probe the OUTx node to check if there is any pulse injected during clipping.

Thanks!

Regards,

Hao

Hi Hao

As far as I can tell Pulses occur during clipping

Below is a capture with one trace on the Outx pin. I don't think the capture sample rate is fast enough to get all the pulses.

The CLIP_OTW pin toggles the Moment these spiking occurs on the output, so I can only surmise that they are related

Thanks

during clipping

Hi,

Yes, pulse injection can be observed clearly. Pulse injection only take effect when there is a clipping condition. It's designed to re-charge the bootstrap cap during high side always ON condition in clipping. If slightly clipping happens, pulse injector may not start to work.

Regards,

Hao

Hi

So back to my original question, this ringing on the waveform is Normal? and how it is very pronounced when has Load

?

Thanks

Brett

Hi,

The last waveform is normal, is there significant audible noise? If there is, then the noise is reasonable. 

What's the PVDD and GVDD you are using? Looks like PVDD is about 8V? I'm thinking that insufficient GVDD voltage could make BST cap not fully charged, which means more pulses injected during clipping and more noise brought due to the injection.

Please apply 12V on both GVDD and VDD, then check again.

Any further question, please let us know. Thanks!

Regards,

Hao

Hi Hao

 GVDD_xx  has always been 12V (checked with measurement) and PVDD_XX is 43v

VDD is 12V

DVDD is 3.3v

AVDD is 7.8v

BVG is 1.25v

The Noise ends up being right up in the Upper end of the Audio Band. like 20 + khz.

The last Picture was No Load and Very Overdriven

Please see Previous pictures for real world into Load noise.

are you able to send me a picture of your board at light clipping to compare

As you can see on my picture During Heaving Clipping the flat top noise is what I expect and looks like the datasheet. But its the Leading and Trailing  noise that concerns me when the clipping amount is lower as it comes out of clipping.

Thanks

Hi

To add some more info

below is a capture under 8 Ohm Load of a 1kz sine wave source

Here you can clearly see the ripple of the Pulse injector during the Flattop Clipping , As the Ripple is 1/4 of the switching Freq or 450khz. This looks normal as expected.

But the leading and trailing edge just before and just after clipping the ringing occurs at a lower frequency

Could this be amp instability?

Hi,

I did some bench test and got the similar result to you.

I guess this might because at the clipping start point, the pulse injection on OUT+ and OUT- legs can have some imbalance, the PWM on two legs is not strictly complementary. So the differential output looks like instability.

Regards,

Hao