TPS63020: How to verify the stability of DCAP mode synchronous buck converter

Muhsiu Wei36

Part Number: TPS63020

Hi Team,

As you know, DCAP(COT) control mode is popular in the system synchronous buck converter now. Network analyzer is used to measure the Bode plot and confirm the converter stability. In the COT control, there is no Error Amplifier in the control loop, only comparator. It's similar with hysteresis control. The question is, Do we really need measure bode plot to verify the DCAP mode converter stability? or we can only use load transient to verify DCAP mode converter stability?

Thank you.

Muhsiu.

over 5 years ago

0 Milos over 5 years ago

TI__Expert 7925 points

Hi Muhsiu,

The TPS63020 has average current mode control, not DCAP. Are you using the TPS63020 or another device?

You can use load transients to check the stability of a DCAP converter. Otherwise, before measuring bode plots, please refer to this article:

https://e2e.ti.com/blogs_/b/powerhouse/archive/2015/05/05/power-tips-how-to-measure-bode-plots-with-dcap-and-dcapx-regulators

Best regards,
Milos

0 Muhsiu Wei36 over 5 years ago in reply to Milos

Intellectual 340 points

Hi Milos,

Sorry for the confusion. I didn't use TPS63020. I saw the discussion of phase margin in the topic (TPS63020: Phase margin improvement method). And, I think my question is related with phase margin, so i click the "Ask a related question". May i continue ask question on this post? or i need to create new one?

Thanks for sharing the article. If i only use load transient to verify DCAP converter. What is the test condition i should follow?(For example: 50% load step or 100% load step ; slew rate? ; output voltage pass/fail criteria? ) Do you have any guideline can refer?
Thank you.

Muhsiu.

0 Milos over 5 years ago in reply to Muhsiu Wei36

TI__Expert 7925 points

Hi Muhsiu,

I would use a load step from zero up to the output current I expect in my application. The slew rate should be faster than the bandwidth of the converter. Typically for our devices a 1 μs rise/fall time is fast enough. Otherwise, if you use very fast rise/fall times, make sure you know what is the control response and what is caused by the output impedance.

http://www.ti.com/lit/an/slyt151/slyt151.pdf

As for pass/fail criteria, obviously if the output oscillates indefinitely or with very little damping, the converter is unstable. If there are just a few oscillations, it is up to you to decide if it is acceptable. Usually, the number of oscillations during the load transient is linked with the amount of over/undershoot, which is also linked to the phase margin. Here is some literature on evaluating stability using the load transient response:

http://www.ti.com/lit/an/slva289b/slva289b.pdf
http://www.ti.com/lit/an/slva381b/slva381b.pdf
http://www.ti.com/lit/an/slyt732a/slyt732a.pdf

I do not cover devices with DCAP control, our buck-boost converters such as the TPS63020 have current-mode control, and DCAP is mostly in our buck converters. If you are specifically interested in DCAP, I suggest finding a fitting DCAP device, opening a new thread, and the expert for DCAP will respond.

Best regards,
Milos

0 Muhsiu Wei36 over 5 years ago in reply to Milos

Intellectual 340 points

Hi Milos,

Thanks for your sharing and explain. I will study those article.

Muhsiu.

0 Milos over 5 years ago in reply to Muhsiu Wei36

TI__Expert 7925 points

Hi Muhsiu,

You're welcome. I will then close this thread. If you have any further questions just post it here, or open a new one if this one gets locked.

Best regards,
Milos

Power management

Power management forum

TPS63020: How to verify the stability of DCAP mode synchronous buck converter