TPS631000: Extra SW node pulse in boost PFM?

Trenton Reed

Part Number: TPS631000
Other Parts Discussed in Thread: TPS63900

Hi Team,

My customer is experimenting with the TPS631000EVM with an input voltage of 2.6V, output 3.3V and load current of ~20mA (160-Ohm load resistor). When they took a measurement at L1 and L2 it showed two close pulses followed by a delay before repeating 2 pulses again. The PFM/PSM in the datasheet says it operates in "single pulse mode to reduce output ripple." What could be causing this additional pulse?

Thank you,

over 2 years ago

0 Tao Zhang over 2 years ago

TI__Mastermind 26380 points

Hi TR,

it is an issue caused by unstable operation internal. may be a feedforward capacitor will help. please take the attachment as ref.

thanks

Regards

Tao

Efficiency comparison test of TPS63805 and TPS631000.pptx

0 Trenton Reed over 2 years ago in reply to Tao Zhang

TI__Intellectual 2960 points

Thanks, Tao. They're evaluating the full range from 1.6Vin to 5.5Vin for 3.3Vout at about 20mA load but real input should be between 2V and 4.5V. Should they still target the 51pF Cff?

+1 Tao Zhang over 2 years ago in reply to Trenton Reed

TI__Mastermind 26380 points

Hi TR,

if their maximum output current is 20mA and PFM working only, i think a 51PF cff should be ok.

but if 20mA is the maximum output current, i think TPS63900 is a better option.

Regards

Tao

0 Trenton Reed over 2 years ago in reply to Tao Zhang

TI__Intellectual 2960 points

Hi Tao,

Can you test the 51pF for stability down to 2V and up to 4.5V? I received the EVM today and tested 47pF myself... looked ok down to 2.8V but then I started to see an additional pulse again.

Bode plots would be great, too. I don't have that capability in my lab.

Thank you,

0 Tao Zhang over 2 years ago in reply to Trenton Reed

TI__Mastermind 26380 points

Hi TR，

sorry i missed you update, and i will show you later.

Regards

Tao

0 Tao Zhang over 2 years ago in reply to Trenton Reed

TI__Mastermind 26380 points

Hi TR,

just kindly reminder, it is unable to test bode plots during PFM working for there will be extra low frequency ripple.and loop is no longer work during PFM working.

Regards

Tao

0 Tao Zhang over 2 years ago in reply to Trenton Reed

TI__Mastermind 26380 points

Hi TR,

Bad news, during my bench test with 51pf, i got same performance with you.

seems it is not a good solution for your application.

so, it is unacceptable for this additional ripple, right?

Regards

Tao

0 Trenton Reed over 2 years ago in reply to Tao Zhang

TI__Intellectual 2960 points

Is there a better value you'd recommend 10pF or 100pF?

The ripple difference when the additional pulse is there in your buck mode slides is pretty substantial. Looks like you also referenced that you tested ripple at lower Vins (boost mode) can you share those, too?

Their load is pretty small on the board <100mA max but it's connected to an IoT device which will make it pretty dynamic. We just want to make sure it's going to be stable.

+1 Tao Zhang over 2 years ago in reply to Trenton Reed

TI__Mastermind 26380 points

Hi TR,

seems we can end disccusion here and begain with mail, right?

Regards

Tao

Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

Power management

Power management forum

TPS631000: Extra SW node pulse in boost PFM?