This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS548D22: Voltage droop when device operating mode switch in no-load start-up condition

Part Number: TPS548D22

Dear team,

We find Voltage droop when device operating mode switch from PFM to FPWM in no-load start-up condition. Waveform as below.

But if we add ~0.8A load at  output and the start-up waveform will be good without voltage drop.

Could you share your comments with us? thanks.

 

Zoom in start up waveform.

Schematic

  • Hello,

    The behavior shown in the scope image is consistent with my expectations for how this part behave. In the attached startup waveform, is it correct to state that a load is applied at the output in order to force continuous switching? 

    If that is the case, the capture indicates that there is an initial dip in the output voltage when the load is first applied, then the output voltage recovers to a lower set point than previously captured under light load operation. During light load operation, the output voltage will regulate incrementally higher when compared to continuous conduction mode as the part exhibits a constant on-time like control mode.

    Regards,

    Alec Biesterfeld

  • Hi Alec,

    yes, we enable output load and enable device.

    If device behavior is the same as your description, output will slightly higher in PFM. And then keep the same voltage level in FPWM.

    I am confusing why output voltage drop once mode switching and then slowly back to high voltage level. During this time, there are no switching mode change and either load change.

    Please share your comments with me. Thanks.

    Regard,

    Ben

  • Hi Ben, 

    I do have a couple of follow up questions:

    1. What is the slew rate of the load (or how long does the load take to ramp up) and what is current level is the load set to when enabled?

    2. Where on the schematic (or board are you measuring) the output voltage?

    Regards,

    Alec Biesterfeld

  • Hi Alec,

    1. There are no load apply at output side when issue happened. 

    2. we measure output voltage at output capacitor close to regulator side.

    Regards,

    Ben

  • Ben,

    Thanks for answering the follow up questions. Based on the capture, it appears that the device sees a load event based on the transition from DCM to FCCM (although there is not a noticeable change in the Iout waveform that may indicate a load is present on the device). During the load event and transition from DCM to FCCM, it is reasonable to expect a few mV deviation in the output voltage setpoint along with a  initial "droop" in the output depending on the speed at which the load is applied.

    As far as troubleshooting this issue further goes, I do recommend probing the output across an output capacitor close to the regulator as you are doing.

    In terms of modifications to the schematic that may help alleviate the issue I recommend the following:

    1. Change R990 to 33.2kohm (for light load efficiency) or 32.4kohm (for FCCM operation). This will change the time constant and subsequent RAMP setting internal to the part. This is recommended for higher duty cycle applications and may help correct the issue your seeing at light load.

    2. I would also recommend that C867 be unpopulated. I don't expect this to be the source of the issue, but if there is any change in the voltage seen by the capacitor, this may affect the internal reference and affect regulation. This pin is specified as do not connect in the datasheet for this reason.

    Regards,

    Alec Biesterfeld