TPS54560: The problem of loud ripple noise

Rhea Gao

Expert 2095 points

Part Number: TPS54560

Hi,

I used TPS54560 test, measured the ripple is very large, at 700mV.

Vin 15-30V, inductively selected 22uH, no-load test, 50mA. The attachment is the schematic diagram and PCB cloth board.

Please help to suggest what the problem is, many thanks~

Design schematic and pcb cloth board.docx

over 1 year ago

0 David Baba over 1 year ago

TI__Mastermind 26350 points

Hello Rhea,

please can you take some measurements of Vswitch as well as Vout AC coupled. Thanks.

David.

0 Rhea Gao over 1 year ago in reply to David Baba

Expert 2095 points

Hi David,

Please kindly tell me where the Vswitch and Vout AC coupled are measured. Many thanks~

0 David Baba over 1 year ago in reply to Rhea Gao

TI__Mastermind 26350 points

Hello Rhea,

Vswitch is at the SW pin of the Device. Measure the output voltage across the output capacitor with an oscilloscope with the coupling set to AC. Thanks.

David.

0 Rhea Gao over 1 year ago in reply to David Baba

Expert 2095 points

Hi David,

I updated the test diagram in the attachment.

In addition, GND no through hole, whether this will have an effect?

4774.Design schematic and pcb cloth board.docx

0 David Baba over 1 year ago in reply to Rhea Gao

TI__Mastermind 26350 points

Hello Rhea,

A couple of things I noted on the measured results you have provided.

The first is that the time base is very long to it is difficult to determine if the ripple is excessive. Reduce time base on scope and you should see that the ripple is in sync with the switch node which looks normal.

You mentioned a ripple of 700mV, however the actual ripple is lower than this, I would say around 200mV peak to peak. The noise spikes you see are much greater than the ripple.

The noise spikes are likely caused by ground pick up on your scope probe lead. Suggest kelvin connect using a spring clip on the barrel of your probe and measure directly across the COUT.

If you wish to reduce the ripple voltage mentioned above to less than 200mV, then suggest reducing the ESR of the output capacitor.

Hope this helps.

David.

Power management

Power management forum

TPS54560: The problem of loud ripple noise