• 状态
  • 回复 5 回复
  • 订户 57 订户
  • 视图 99 视图
  • 用户 0 会员在此处
选项
选项
相关

TPS55289: TPS55289 not working after being merged into main board

Hao Lennart
Prodigy 20 points

Part Number: TPS55289
Other Parts Discussed in Thread: BQ25792, TPS54560

Hi TI team,

I’m running into a strange issue with TPS55289 after merging it into a larger power board.

Power structure:
Adapter → BQ25792 → TPS54560DDAR (5 V rail) → TPS55289 (digitally controlled output)

Here’s the situation:

  • TPS55289 works perfectly on its standalone board

  • BQ25792 and TPS54560 also work fine on their own

  • After merging TPS55289 into the main board, TPS55289 no longer works correctly

For the merged board, the TPS55289 layout was done following the EVM layout guidelines and reference design, and is basically identical to the standalone board.
The inductor (L2) is the same part as used on the reference design.

What I see on the merged board

  • After writing registers to set the output to 5 V

  • VOUT ramps up to about 4.5 V, then drops back to around 3.8 V

  • Then it ramps up again to ~4.5 V and repeats this cycle

  • The inductor makes an audible hiccup / clicking / popping sound

  • SW is switching, but not continuously

  • Measured BOOT–SW ≈ 4.5 V, so bootstrap looks OK

It looks like the converter is trying to start, then backing off repeatedly.

Key observation

If I take the TPS55289 standalone board and connect it into the system with flying wires (powered from the same 5-V rail), it works normally and stably.

That’s why I’m suspecting something subtle related to layout, grounding, return paths, or interaction with the rest of the system, even though the TPS55289 section itself looks the same as the small board.

Has anyone seen similar behavior with TPS55289, or have ideas on what might cause this kind of repeated ramp-up / drop-back after merging into a larger board?

Thanks!

image.pngimage.pngimage.pngimage.pngimage.pngimage.png

  • 0
    TI__Mastermind 34145 points

    Hello, we have received your case and the investigation will take some time. Thank you for your patience.

  • 0
    TI__Mastermind 44485 points

    HI

    I think the issue was caused by upstream circuit of TPS55289 -TPS54560DDAR (5 V rail).

    The layout of TPS54560DDAR is poor and output cap of ceramic is also not sufficient.

    First, add 22uFx2 ceramic cap at buck output. If not work, please put the GND of power diode and output cap like below shown.

  • 0 回复 Daniel
    Prodigy 20 points

    Hi,

    Thanks for your feedback. However, I would like to clarify a key point: During my testing on the merged board, the TPS54560DDAR was not populated. I provided the 5V rail directly from a stable, high-quality external DC power supply.

    Despite having a stable input source, the TPS55289 on the merged board still exhibits the same behavior: VOUT ramps to 4.5V, drops to 3.8V, and the inductor makes a "hiccup" sound. Under the exact same setup (using the same external supply and flying wires), the standalone board works perfectly.

    Since the upstream converter (TPS54560) is ruled out, could there be other factors related to the larger board's system ground plane, feedback routing interference, or output parasitic parameters that might trigger this repeated restart?

  • 0 回复 Hao Lennart
    TI__Mastermind 44485 points

    Hi

    Please use the short and thick wire for connection from external DC power supply to TPS55289.

    Also, please provide the input, SW1, SW2, Vout waveforms for check.

  • 0 回复 Daniel
    Prodigy 20 points

    Hi Daniel,

    Sorry for the delayed response — I just came back from a short holiday.

    Regarding your suggestion, I have already tried using short and thick wires to connect the external DC power supply to the TPS55289 VIN and GND, but the behavior remains unchanged on the merged board.

    The SW waveforms are attached below (placed after this reply for reference).
    As for VOUT, it is not stable — it shows strong up-and-down oscillation, repeatedly ramping up and collapsing, which matches the audible hiccup sound from the inductor.

    Under the exact same external supply and wiring setup, the standalone TPS55289 board operates normally. This still suggests the issue is related to the merged board implementation rather than the input source.