LM5175: Help with Output Ripple in LM5175 Buck-Boost Circuit

Sadaharu Kanda

Part Number: LM5175

Tool/software:

Hi there,

I'm working on a buck-boost converter using the LM5175 with the following specs:
Input: 4.5-16V
Output: 5V/10A max
Switching frequency: 500kHz

I'm seeing an output ripple of about 100-150 mVpp at around 20 kHz, and I'm having a hard time figuring out what's causing it.
Could you give me some pointers on what to check, or suggest any changes to the component values that might help?

Here's a summary of the main components I'm using:
Cout: 10uF x9 (MLCC), 4.7µFx2 (MLCC)
Cin: 10uF x6 (MLCC)
L: 1uH
Rs: 5 m
Rt: 48.7 k
Css: 0.1uF
Cslope: 56 pF
Rc: 2 k
CC1: 4700 pF
CC2: 330 pF
Mode: 93.1 k (CCW / Hiccup enabled)

If there's an upload link or email address, I can send over the schematic and some scope waveforms too.

Thanks,
Kanda

1 month ago

0 Stefan Schauer 1 month ago

TI__Guru 58641 points

HI Kanda,

you can attach the files here - by drag and drop.

If you would like to send them via private channel you can hover over my icon and select send private message.
(But not that this can then only be seen by me and also no colleague can answer in that case.)

When you provide the scope plots with the measurement please also add detailed description on your measurement setup - best with pictures)

Best regards,

Stefan

0 Sadaharu Kanda 1 month ago in reply to Stefan Schauer

Prodigy 10 points

Stefan,

Please refrain from publishing this for the sake of my customers. I would like to send you a direct message, but when I click on your icon, it takes me to the attached profile screen, and I don't know what to do next.

Thanks,

S.Kanda

0 Stefan Schauer 1 month ago in reply to Sadaharu Kanda

TI__Guru 58641 points

HI Kanda,

If you would like to send them via private channel you can hover over my icon (Not Click) and select send private message.

Best regards,

Stefan

0 Sadaharu Kanda 1 month ago in reply to Stefan Schauer

Prodigy 10 points

Hi Stefan,

Thank you for your prompt response.

For now, I'm sending you documents by direct message.

And now,We have identified a phenomenon that suggests fluctuations in the load-side current may be the cause of the issue.

I'll follow up with additional materials later.

Best regards,

S.Kanda

0 Stefan Schauer 1 month ago in reply to Sadaharu Kanda

TI__Guru 58641 points

HI Kanda,

Could you please attach the filled out LM5175QUICKSTART-CALC to this thread? This would be the most efficient workflow for us to do a schematic review.

It might be that the compensation is not stable for your design.

Best regards,

Stefan

0 Sadaharu Kanda 1 month ago in reply to Stefan Schauer

Prodigy 10 points

Hi Stefan,

Attached is the Excel file you requested.
However, I’m sorry to say that the section related to efficiency has not been edited yet.

Thank you,
S. Kanda

0 Stefan Schauer 1 month ago in reply to Sadaharu Kanda

TI__Guru 58641 points

HI Kanda,

no problem about the efficiency section.

You have entered 100uF for the output cap. I have you used ceramic caps for the 10uF and 4.7uF - if yes, have you checked the DC BIAS derating.

The compensation values in the excel and the schematic do not match -can you check and clarify that.

Best regards,

Stefan

0 Sadaharu Kanda 1 month ago in reply to Stefan Schauer

Prodigy 10 points

Hi Stefan,

Here’s the updated Excel file that takes into account the capacitor variation you pointed out.
I’ve also attached the capacitor catalog.

By the way, where in the Excel should I look to find the recommended values for the phase compensation components?
I'd like to test them on the real circuit.

Best regards,
S. KandaCOUTIN.xlsx

0 Stefan Schauer 1 month ago in reply to Sadaharu Kanda

TI__Guru 58641 points

Hi Kanda,

can you please try to increase the phase margin to ~ 60 Degree.

45 is very low - esp. when this is only on model basis. Even when measured on the board I would not go that low.

Also set the Crossover Frequency not high then the low of:

- f_SW / 5 (better / 10)

- RHPZ (right have plane zero) / 3 ( better / 5)

So in your design ~ 10k

and this COMP setting could work:

Best regards,

Stefan

0 Sadaharu Kanda 1 month ago in reply to Stefan Schauer

Prodigy 10 points

WBDesign76_5-16VIN_5V10A_500KHz_1uH.pdf

Hi Stefan,

We are currently trying out the compensation circuit you suggested.
Attached is the WEBENCH report that served as the basis for our component values. (It took us a while to locate it.)

On page 11, item 70 shows a phase margin of 45 degrees — do you still think this needs to be changed?

Best regards,
Kanda

0 Stefan Schauer 1 month ago in reply to Sadaharu Kanda

TI__Guru 58641 points

HI Kanda,

i recommend to have a:

- phase margin of > 60 Degree

- gain margin of > -12db

Esp. if this is only based on calculated values and not measured on the real PCB.

Best regards,

Stefan

+1 Sadaharu Kanda 1 month ago in reply to Stefan Schauer

Prodigy 10 points

Hi Stefan,

Understood.
Thank you for your prompt response.

Best regards,
S. Kanda

Power management

Power management forum

LM5175: Help with Output Ripple in LM5175 Buck-Boost Circuit