TPS552892: Vout abnormal

Jeff YANG

Part Number: TPS552892

Tool/software:

Hi,

We are using a TI buck-boost IC, TPS552892, and have encountered an issue with unstable power compensation, as shown in <Figure 1> and <Figure 2>. Ultimately, due to excessive peak current triggering the protection component, the power source was disconnected.
The circuit diagram is shown in <Figure 3>, and it is essentially configured according to the EVM setup.
The input and output voltages are both 12V, with a current transient of approximately 100mA to 1.5A.

Have you encountered this issue before, and what might be the possible causes?
Would increasing the switching frequency (Fsw) to improve response speed effectively stabilize the output voltage in a timely manner?
We are now planning to adjust the Fsw frequency to 1.2MHz (15kΩ). For the frequency response, do you have any recommended values for the components R1440/R1568/C1569/L19 in <Figure 3>?
I have tried to using webebch tool to design for 1.2Mhz, but why only can choose 0.2 - 1.067Mhz? Is it webebch tool bug?

Figure 1

Figure 2

Figure 3

Thanks!

Jeff

3 months ago

0 Mounika Girish 3 months ago

TI__Intellectual 2426 points

Hello Jeff,

Thank you for using e2e forum,

Our experts of this device is out of office. Please expect a response early next week. Thank you for your patience.

Best regards,

Mounika

0 Jeff YANG 3 months ago in reply to Mounika Girish

Mastermind 7825 points

Hi,

I've tried to using webench tool generate below schematic, but I use this schematic put in excel caculate tool, the BW only have ~4.5kHz at 0dB.

Should I adjust the gain @0dB to a higher frequency? And should it be based on the frequency of the load changes?

And keep phase margin >45~60 degree?

TPS552892.pdf

TPS55289-CALCULATION-TOOL-V1.0 的複本.xlsx

Thanks!

Jeff

0 Jeff YANG 3 months ago in reply to Jeff YANG

Mastermind 7825 points

Hi,

Another question is the f(BW) desired, if I want set higher BW to get better transient reponse, the limitation should be "less than 1/5*fzRPH and 1/10*fsw", right?

Does there have any other IC internal limitaion?

Thanks!

Jeff

0 Mulin Yuan 3 months ago in reply to Jeff YANG

TI__Genius 14950 points

Hi Jeff,

What is input voltage range, output voltage, output current? What is the part number of output caps?

Regards,

Mulin

0 Jeff YANG 3 months ago in reply to Mulin Yuan

Mastermind 7825 points

Hi Mulin,

Vin in normal conditon is 10.8V ~ 13.2V, When power off protection will drop to 7V.

Output cap is YAGEO CC0805MKX5R8BB226.

Please help f(BW) design limitaion "less than 1/5*fzRPH and 1/10*fsw"soon. Or have other limitation we need to noticed?

We also found TPS552892EVM in vin 12V to vout 12V output voltage sometimes not stable, does EVM default compensetion value is okay for 12V to vout 12V design?

thanks!

Jeff

0 Jeff YANG 3 months ago in reply to Jeff YANG

Mastermind 7825 points

Hi,

Just summerize above questions as below.

1. What parameters should we consider in our design to ensure there are no issues? For example: crossover frequency, phase margin, gain margin, Q, damping?(For Vin =10.8V(7V) ~ 13.2V, When power off protection will drop to 7V, Vout=12V)

2. When the parameter R8 in the original EVM design is adjusted from 15kΩ to 20kΩ, the output voltage waveform (blue section in the figure below) starts to exhibit oscillation (without load). Is it normal?

Thanks!

Jeff

0 Shipeng Cheng 3 months ago in reply to Jeff YANG

TI__Genius 17290 points

Hi Jeff,

Your waveform shows the loop is unstable. So you need to change to smaller Rcomp and larger Ccomp. Usually the bandwidth of loop is less than 10kHz. Please fill in the calculation tool with correct value in your circuit, for example L, Cout, ESR and compensation value. Then you can get correct bode plot in the tool. Thanks.