TPS566231: Webench efficiency plot

Christian Lieber

Part Number: TPS566231

Hi TI-experts,

I generated a TPS566231 design using the webench designer.

In the efficiency graph (please see attachment) I found a working point (in-between two high efficiency points) that has a much lower efficiency. (Vout = 0.95V)

Does this reflect the real behavior or this this just an simulation issue?

Best regards,
Christian

over 3 years ago

0 Miranda Gu over 3 years ago

TI__Genius 15327 points

Hi Christian,

Thanks for your question!

The Webench simulation may not reflect real efficiency performance.

I checked our validation report, for TPS566231, efficiency became higher when Vin became lower at light load (<0.1A).

Let me know if you have further question.

0 Christian Lieber over 3 years ago

Prodigy 30 points

Christian Lieber said:

Thaks for the response.

Please let me recap to make sure my understanding is right. The efficiency increases with decreasing input voltage, without any local minima.

Thus in the real world, the efficiency for 3.675V will be between the ones for 4.35 V and 3.0 V?

Best regards,
Christian

+1 Miranda Gu over 3 years ago in reply to Christian Lieber

TI__Genius 15327 points

Hi Christian,

We didn't measure efficiency at 3.675Vi, since it's not a typical voltage.

While we measured efficiency at 3Vi/5Vi/8Vi/12Vi/18Vi 1Vo@light-load(<0.1A) conditions. The result was that efficiency increased with Vi decreased, 3Vi condition got the highest efficiency.

0 Christian Lieber over 3 years ago in reply to Miranda Gu

Prodigy 30 points

Hi Miranda,

I totally understand that you can't measure every voltage point.

But do you expect the efficiency to drop in between, due to any state changes inside the die? Or do you think it's due to modeling inaccuracy inside the design tool?

In the data sheet I can't find any evidence for this behavior. But I wanted to make sure I didn't miss any detail on this.

Best regards,
Christian

0 Ruby Li over 3 years ago in reply to Christian Lieber

TI__Expert 7430 points

Hi Christian

Miranda is looking into this and will feedback to you soon. Thanks.

Ruby

0 Miranda Gu over 3 years ago in reply to Christian Lieber

TI__Genius 15327 points

Hi Christian,

Agree with you. I think it's a modeling error.

Power management

Power management forum

TPS566231: Webench efficiency plot